resume-screening-system

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,29 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+dist/
+build/
+*.egg
+.pytest_cache/
+
+# Virtual environment
+venv/
+env/
+.venv/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# Uploads (runtime data)
+uploads/
+
+# OS files
+.DS_Store
+Thumbs.db
+
+# Environment variables
+.env

Dockerfile

@@ -0,0 +1,34 @@
+FROM python:3.9-slim
+
+WORKDIR /app
+
+# Install system dependencies — ffmpeg required for Whisper audio processing
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    ffmpeg \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first (for Docker layer caching)
+COPY requirements.txt .
+
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Download NLTK data (needed for text processing)
+RUN python -c "import nltk; nltk.download('stopwords'); nltk.download('punkt'); nltk.download('punkt_tab')" 2>/dev/null || true
+
+# Copy all project files
+COPY . .
+
+# Create upload directories
+RUN mkdir -p uploads/resumes uploads/audio
+
+# Expose port required by Hugging Face Spaces (Docker SDK)
+EXPOSE 7860
+
+# Environment variables for Flask
+ENV FLASK_RUN_PORT=7860
+ENV FLASK_RUN_HOST=0.0.0.0
+ENV PYTHONUNBUFFERED=1
+
+# Production server with gunicorn — 120s timeout for ML model loading + Whisper
+CMD ["gunicorn", "-b", "0.0.0.0:7860", "app:app", "--timeout", "120", "--workers", "2", "--threads", "4"]

README.md

@@ -0,0 +1,179 @@
+---
+title: HireScope AI
+emoji: 🔍
+colorFrom: indigo
+colorTo: cyan
+sdk: docker
+app_file: app.py
+python_version: "3.9"
+pinned: false
+---
+
+# 🔍 HireScope AI — Intelligent Resume Screening
+
+An AI-powered web application that parses resumes, extracts 250+ skills, matches candidates to job descriptions using **Sentence-Transformers** semantic embeddings, ranks them with hybrid scoring, and transcribes audio intros with **OpenAI Whisper** — built with Flask, MongoDB Atlas, Cloudinary, and Tailwind CSS.
+
+![Python](https://img.shields.io/badge/Python-3.9+-blue)
+![Flask](https://img.shields.io/badge/Flask-3.x-green)
+![Whisper](https://img.shields.io/badge/Whisper-OpenAI-orange)
+![MongoDB](https://img.shields.io/badge/MongoDB-Atlas-green)
+![Tailwind](https://img.shields.io/badge/Tailwind-CSS_v3-blue)
+
+---
+
+## ✨ Features
+
+| Feature | Description |
+|---|---|
+| **Resume Upload** | Upload PDF or DOCX files for automatic text extraction (including tables) |
+| **250+ Skill Extraction** | Regex-based extraction across 24 BTech categories (Software, ML, DevOps, Embedded, etc.) |
+| **Skill Normalization** | Infers higher-level skills (e.g., "TensorFlow" → "Deep Learning" added) |
+| **Semantic Matching** | Sentence-Transformer (`all-MiniLM-L6-v2`) embeddings + Cosine Similarity |
+| **Hybrid Scoring** | 50% Semantic Score + 50% Exact Skill Overlap → Score 0–100 |
+| **Skill Gap Analysis** | Shows matched ✅ and missing ❌ skills for each candidate |
+| **Audio Transcription** | Upload voice recordings → transcribed via OpenAI Whisper (local-file-first approach) |
+| **Async Processing** | Audio transcription runs in background thread via ThreadPoolExecutor |
+| **AJAX Polling** | Real-time transcription status updates without page reload |
+| **Candidate Ranking** | Leaderboard with re-ranking via custom JD |
+| **Candidate Profiles** | Click any candidate → full modal with skills, education, experience, audio |
+| **Resume Download** | Cloudinary-hosted with `fl_attachment` for direct download |
+| **Auth System** | Login/Register with Werkzeug password hashing |
+| **Premium UI** | Tailwind CSS light theme + Inter font + glassmorphism + animations |
+
+---
+
+## 🗂️ Project Structure
+
+```
+HireScope-AI/
+├── app.py                    # Flask main app (routes, auth, APIs)
+├── resume_parser.py          # PDF/DOCX text extraction + name/email/phone extraction
+├── skill_extractor.py        # 250+ skills regex extraction + normalization
+├── job_matcher.py            # Sentence-Transformers matching + ranking engine
+├── audio_transcriber.py      # Whisper speech-to-text (local-file-first approach)
+├── db.py                     # MongoDB CRUD operations
+├── requirements.txt          # Python dependencies
+├── Dockerfile                # Docker deployment config (HuggingFace Spaces)
+├── .env                      # Sensitive credentials (NOT in Git)
+├── docs/
+│   ├── SRS.md                # Software Requirements Specification
+│   ├── JIRA_PLAN.md          # Sprint planning & bug tracker
+│   └── REVISION.md           # Viva revision guide (Hinglish + English)
+├── templates/
+│   ├── base.html             # Shared layout (nav, footer, flash messages)
+│   ├── login.html            # Auth - sign in
+│   ├── register.html         # Auth - sign up
+│   ├── index.html            # Dashboard + resume upload (drag & drop)
+│   ├── results.html          # Analysis results + audio upload + transcription
+│   └── ranking.html          # Candidate leaderboard + profile modal
+├── static/css/style.css      # Supplemental CSS
+└── uploads/
+    ├── resumes/              # Temporary resume storage
+    └── audio/                # Temporary audio storage (for Whisper)
+```
+
+---
+
+## 🚀 How to Run
+
+### Prerequisites
+
+- **Python 3.9+** installed
+- **ffmpeg** installed (required for Whisper audio processing)
+  - Windows: `winget install ffmpeg`
+  - Mac: `brew install ffmpeg`
+  - Linux: `sudo apt install ffmpeg`
+
+### Step-by-Step Setup
+
+```bash
+# 1. Clone the repository
+git clone https://github.com/yourusername/HireScope-AI.git
+cd HireScope-AI
+
+# 2. Create a virtual environment
+python -m venv venv
+
+# 3. Activate the virtual environment
+# Windows:
+venv\Scripts\activate
+# Mac/Linux:
+source venv/bin/activate
+
+# 4. Install dependencies
+pip install -r requirements.txt
+
+# 5. Download NLTK data (one-time)
+python -c "import nltk; nltk.download('stopwords'); nltk.download('punkt')"
+
+# 6. Create .env file with your credentials
+# MONGO_URI=mongodb+srv://...
+# CLOUDINARY_CLOUD_NAME=...
+# CLOUDINARY_API_KEY=...
+# CLOUDINARY_API_SECRET=...
+# GOOGLE_API_KEY=...  (optional)
+# SECRET_KEY=your-secret-key
+
+# 7. Run the application
+python app.py
+```
+
+### Open in Browser
+
+```
+http://127.0.0.1:5000
+```
+
+---
+
+## 📖 How It Works
+
+### 1. Resume Parsing (`resume_parser.py`)
+- **PDF** files parsed using `PyPDF2`
+- **DOCX** files parsed using `python-docx` (paragraphs + tables)
+- Extracts candidate name, email, phone from raw text
+
+### 2. Skill Extraction (`skill_extractor.py`)
+- **250+ skills** across 24 categories covering all BTech career paths
+- Regex-based **word-boundary matching** for precision
+- **Skill normalization** infers related skills
+
+### 3. Semantic Matching (`job_matcher.py`)
+```
+Resume Text  →  Sentence Embedding (384-dim)  ─┐
+                                                ├→  50% Cosine Similarity + 50% Skill Overlap → Score (0-100)
+Job Description  →  Sentence Embedding (384-dim) ─┘
+```
+
+### 4. Audio Transcription (`audio_transcriber.py`)
+- Uses OpenAI's Whisper (`base` model)
+- **Local-file-first approach** — audio saved to disk, transcribed from local path
+- Supports MP3, WAV, M4A, FLAC, OGG, WEBM
+- Runs asynchronously via `ThreadPoolExecutor`
+
+### 5. Candidate Ranking
+- Leaderboard with re-ranking via custom JD
+- Clickable candidate profiles with full biodata modal
+
+---
+
+## 🛠️ Tech Stack
+
+| Component | Technology |
+|---|---|
+| Backend | Python 3.9, Flask 3.x |
+| Text Extraction | PyPDF2, python-docx |
+| NLP | Regex (250+ skills), NLTK |
+| Semantic ML | Sentence-Transformers (`all-MiniLM-L6-v2`) |
+| Audio ASR | OpenAI Whisper (base model) |
+| Database | MongoDB Atlas (PyMongo) |
+| File Storage | Cloudinary |
+| Frontend | Tailwind CSS v3, Inter Font, Vanilla JS, AJAX |
+| Auth | Werkzeug password hashing, Flask sessions |
+| Deployment | Docker, Gunicorn, HuggingFace Spaces |
+
+---
+
+## 📄 License
+
+This project is for educational purposes. Feel free to use and modify.

app.py

@@ -0,0 +1,484 @@
+"""
+app.py
+------
+Main Flask application for HireScope AI — Resume Screening System.
+Includes MongoDB, Auth, Async Processing, and Sentence Transformers.
+"""
+
+import os
+import logging
+import concurrent.futures
+from functools import wraps
+from flask import (
+    Flask, render_template, request, redirect, url_for,
+    flash, session, jsonify
+)
+from werkzeug.middleware.proxy_fix import ProxyFix
+from dotenv import load_dotenv
+import cloudinary
+import cloudinary.uploader
+import cloudinary.api
+
+# Load environment variables
+load_dotenv()
+
+# Cloudinary Configuration
+cloudinary.config(
+    cloud_name=os.getenv("CLOUDINARY_CLOUD_NAME"),
+    api_key=os.getenv("CLOUDINARY_API_KEY"),
+    api_secret=os.getenv("CLOUDINARY_API_SECRET"),
+    secure=True,
+)
+
+from db import (
+    create_user, authenticate_user, insert_candidate, 
+    get_all_candidates, get_candidate_by_id, update_candidate_audio,
+    update_candidate_audio_error, set_candidate_audio_processing,
+    clear_all_candidates
+)
+from werkzeug.utils import secure_filename
+from resume_parser import extract_text, clean_text
+from skill_extractor import extract_all, SKILLS_LIST
+from job_matcher import calculate_match_score, find_skill_gaps, rank_candidates
+from audio_transcriber import transcribe_from_local_file
+
+app = Flask(__name__)
+logging.basicConfig(
+    level=os.getenv("LOG_LEVEL", "INFO"),
+    format="%(asctime)s %(levelname)s [%(name)s] %(message)s",
+)
+logger = logging.getLogger(__name__)
+
+# Secret key
+app.secret_key = os.getenv("SECRET_KEY")
+
+# Hugging Face / Proxy Configuration
+app.wsgi_app = ProxyFix(app.wsgi_app, x_for=1, x_proto=1, x_host=1, x_prefix=1)
+
+# Session Configuration for iframe compatibility (Hugging Face)
+app.config.update(
+    SESSION_COOKIE_SECURE=True,
+    SESSION_COOKIE_SAMESITE='None',
+    SESSION_COOKIE_HTTPONLY=True,
+)
+
+# Optional: Initialize Google Generative AI if key is present
+GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY", "")
+if GOOGLE_API_KEY:
+    try:
+        import google.generativeai as genai
+        genai.configure(api_key=GOOGLE_API_KEY)
+    except Exception:
+        pass
+
+BASE_DIR = os.path.abspath(os.path.dirname(__file__))
+UPLOAD_FOLDER_RESUMES = os.path.join(BASE_DIR, "uploads", "resumes")
+UPLOAD_FOLDER_AUDIO = os.path.join(BASE_DIR, "uploads", "audio")
+os.makedirs(UPLOAD_FOLDER_RESUMES, exist_ok=True)
+os.makedirs(UPLOAD_FOLDER_AUDIO, exist_ok=True)
+
+ALLOWED_RESUME_EXTENSIONS = {"pdf", "docx"}
+ALLOWED_AUDIO_EXTENSIONS = {"mp3", "wav", "m4a", "flac", "ogg", "webm"}
+
+# Thread pool for async audio transcription
+executor = concurrent.futures.ThreadPoolExecutor(max_workers=2)
+
+def allowed_file(filename, allowed_extensions):
+    return "." in filename and filename.rsplit(".", 1)[1].lower() in allowed_extensions
+
+# ── Authentication Helper ──
+def login_required(f):
+    @wraps(f)
+    def decorated_function(*args, **kwargs):
+        if "user_id" not in session:
+            flash("Please log in to access this page.", "warning")
+            return redirect(url_for("login"))
+        return f(*args, **kwargs)
+    return decorated_function
+
+@app.route("/login", methods=["GET", "POST"])
+def login():
+    if request.method == "POST":
+        email = request.form.get("email")
+        password = request.form.get("password")
+        user = authenticate_user(email, password)
+        if user:
+            session["user_id"] = user["_id"]
+            session["username"] = user["username"]
+            session["role"] = user["role"]
+            flash("Logged in successfully!", "success")
+            return redirect(url_for("index"))
+        else:
+            flash("Invalid email or password", "error")
+    return render_template("login.html")
+
+@app.route("/register", methods=["GET", "POST"])
+def register():
+    if request.method == "POST":
+        username = request.form.get("username")
+        email = request.form.get("email")
+        password = request.form.get("password")
+        success, msg = create_user(username, email, password)
+        if success:
+            flash("Registration successful. Please login.", "success")
+            return redirect(url_for("login"))
+        else:
+            flash(msg, "error")
+    return render_template("register.html")
+
+@app.route("/logout")
+def logout():
+    session.clear()
+    flash("Logged out successfully.", "info")
+    return redirect(url_for("login"))
+
+@app.route("/")
+@login_required
+def index():
+    candidates = get_all_candidates()
+    # Calculate stats
+    avg_score = 0
+    if candidates:
+        scores = [c.get("match_score", 0) for c in candidates]
+        avg_score = round(sum(scores) / len(scores), 1)
+    return render_template(
+        "index.html",
+        candidate_count=len(candidates),
+        avg_score=avg_score,
+        recent_candidates=candidates[:5]
+    )
+
+@app.route("/upload", methods=["POST"])
+@login_required
+def upload_resume():
+    if "resume" not in request.files:
+        flash("No file selected.", "error")
+        return redirect(url_for("index"))
+
+    file = request.files["resume"]
+    if file.filename == "":
+        flash("No file selected.", "error")
+        return redirect(url_for("index"))
+
+    if not allowed_file(file.filename, ALLOWED_RESUME_EXTENSIONS):
+        flash("Invalid file type. Please upload PDF or DOCX.", "error")
+        return redirect(url_for("index"))
+
+    filename = secure_filename(file.filename)
+    if not filename:
+        filename = "resume_file"
+    filepath = os.path.join(UPLOAD_FOLDER_RESUMES, filename)
+    file.save(filepath)
+
+    raw_text = extract_text(filepath)
+    if not raw_text.strip():
+        flash("Could not extract text. Please ensure the PDF/DOCX is not just scanned images.", "error")
+        return redirect(url_for("index"))
+
+    cleaned_text = clean_text(raw_text)
+    extracted_info = extract_all(cleaned_text)
+
+    job_description = request.form.get("job_description", "").strip()
+    match_score = 0.0
+    skill_gaps = {"matched": [], "missing": []}
+    jd_skills = []
+
+    if job_description:
+        from skill_extractor import extract_skills
+        jd_skills = extract_skills(job_description)
+        match_score = calculate_match_score(cleaned_text, job_description, extracted_info["skills"], jd_skills)
+        skill_gaps = find_skill_gaps(extracted_info["skills"], job_description, SKILLS_LIST, jd_skills)
+
+    # --- Cloudinary Upload Resume ---
+    resume_url = ""
+    try:
+        cloudinary_response = cloudinary.uploader.upload(
+            filepath,
+            resource_type="auto",
+            folder="resume_screener/resumes",
+            use_filename=True,
+            unique_filename=True,
+        )
+        # Simply use the secure_url provided by Cloudinary
+        resume_url = cloudinary_response.get("secure_url", "")
+    except Exception as e:
+        logger.error("Cloudinary upload failed: %s", e)
+        resume_url = ""
+
+    # Clean up local file after processing
+    try:
+        os.remove(filepath)
+    except Exception:
+        pass
+
+    # AI Summary using Google Gen AI (if configured)
+    ai_summary = ""
+    if GOOGLE_API_KEY:
+        try:
+            import google.generativeai as genai
+            model = genai.GenerativeModel('gemini-flash-latest')
+            prompt = f"Summarize this candidate in 2 to 3 short sentences emphasizing their top skills, experience, and education based on this resume text:\n{cleaned_text[:3000]}"
+            response = model.generate_content(prompt)
+            ai_summary = response.text.strip()
+        except Exception as e:
+            logger.error(f"Generative AI Error (Summary): {e}")
+
+    candidate_name = os.path.splitext(filename)[0].replace("_", " ").replace("-", " ").title()
+    candidate_data = {
+        "name": candidate_name,
+        "filename": filename,
+        "resume_url": resume_url,
+        "resume_text": cleaned_text,
+        "raw_text_preview": raw_text[:500],
+        "ai_summary": ai_summary,
+        "skills": extracted_info["skills"],
+        "education": extracted_info["education"],
+        "experience": extracted_info["experience"],
+        "match_score": match_score,
+        "skill_gaps": skill_gaps,
+        "job_description": job_description,
+        "audio_transcription": None,
+        "uploaded_by": session["user_id"]
+    }
+    
+    # Save to MongoDB
+    candidate_id = insert_candidate(candidate_data)
+    session["last_candidate_id"] = str(candidate_id)
+
+    flash("Resume analyzed successfully!", "success")
+    return redirect(url_for("results"))
+
+
+def process_audio_local(candidate_id, local_audio_path, audio_url):
+    """
+    Process audio transcription from a LOCAL file (not URL).
+    This avoids Cloudinary download issues entirely.
+    The audio is saved locally first, transcribed, then cleaned up.
+    """
+    logger.info("Starting LOCAL transcription for candidate_id=%s, file=%s", candidate_id, local_audio_path)
+    result = transcribe_from_local_file(local_audio_path)
+    
+    if result["success"]:
+        update_candidate_audio(candidate_id, result["text"], result["language"], audio_url)
+        logger.info("Transcription saved for candidate_id=%s", candidate_id)
+    else:
+        update_candidate_audio_error(candidate_id, result["error"], audio_url)
+        logger.error("Transcription failed for candidate_id=%s: %s", candidate_id, result["error"])
+
+    # Clean up local audio file after transcription
+    try:
+        if local_audio_path and os.path.exists(local_audio_path):
+            os.remove(local_audio_path)
+            logger.info("Cleaned up local audio: %s", local_audio_path)
+    except Exception:
+        pass
+
+
+def _handle_transcription_future(future, candidate_id, audio_url):
+    exc = future.exception()
+    if exc is None:
+        return
+    error_msg = f"Background transcription crashed: {exc}"
+    logger.exception("Unhandled transcription error for candidate_id=%s", candidate_id)
+    update_candidate_audio_error(candidate_id, error_msg, audio_url)
+
+@app.route("/upload_audio", methods=["POST"])
+@login_required
+def upload_audio():
+    # Get candidate_id from form (sent from results page) or session
+    candidate_id = request.form.get("candidate_id") or session.get("last_candidate_id")
+
+    if not candidate_id:
+        flash("Please upload and analyze a resume first before attaching audio.", "error")
+        return redirect(url_for("index"))
+
+    candidate = get_candidate_by_id(candidate_id)
+    if not candidate:
+        flash("Candidate not found. Please upload a resume first.", "error")
+        return redirect(url_for("index"))
+
+    if "audio" not in request.files:
+        flash("No audio file selected.", "error")
+        return redirect(url_for("results"))
+
+    file = request.files["audio"]
+    if file.filename == "":
+        flash("No audio file selected.", "error")
+        return redirect(url_for("results"))
+
+    if not allowed_file(file.filename, ALLOWED_AUDIO_EXTENSIONS):
+        flash("Invalid audio format. Supported: MP3, WAV, M4A, FLAC, OGG, WEBM", "error")
+        return redirect(url_for("results"))
+
+    filename = secure_filename(file.filename) or "audio_file"
+    ext = os.path.splitext(filename)[1].lower()
+    if not ext:
+        ext = ".mp3"
+    
+    # === KEY FIX: Save audio LOCALLY first, then transcribe from local file ===
+    local_audio_path = os.path.join(UPLOAD_FOLDER_AUDIO, f"{candidate_id}_{filename}")
+    file.save(local_audio_path)
+    logger.info("Audio saved locally at: %s (%d bytes)", local_audio_path, os.path.getsize(local_audio_path))
+
+    # Upload to Cloudinary for storage (non-blocking for transcription)
+    audio_url = ""
+    try:
+        cloudinary_response = cloudinary.uploader.upload(
+            local_audio_path,
+            resource_type="video",
+            folder="resume_screener/audio",
+            public_id=f"{candidate_id}_{os.path.splitext(filename)[0]}",
+            use_filename=False,
+            overwrite=True,
+        )
+        audio_url = cloudinary_response.get("secure_url", "")
+    except Exception as exc:
+        logger.warning("Audio Cloudinary upload failed (will still transcribe locally): %s", exc)
+        audio_url = ""  # Not critical — transcription uses local file
+
+    set_candidate_audio_processing(candidate_id, audio_url)
+    
+    # === Transcribe from LOCAL file (not from Cloudinary URL) ===
+    try:
+        future = executor.submit(process_audio_local, candidate_id, local_audio_path, audio_url)
+        future.add_done_callback(
+            lambda f, cid=candidate_id, aurl=audio_url: _handle_transcription_future(f, cid, aurl)
+        )
+    except Exception as exc:
+        error_msg = f"Failed to queue transcription task: {exc}"
+        logger.exception(error_msg)
+        update_candidate_audio_error(candidate_id, error_msg, audio_url)
+        flash(error_msg, "error")
+        return redirect(url_for("results"))
+
+    session["last_candidate_id"] = str(candidate_id)
+    session["awaiting_transcription_for"] = str(candidate["_id"])
+    logger.info("Queued LOCAL transcription for candidate_id=%s", candidate_id)
+    flash("Audio uploaded successfully! Transcription is processing in the background.", "info")
+    return redirect(url_for("results"))
+
+@app.route("/results")
+@login_required
+def results():
+    candidate_id = session.get("last_candidate_id")
+    candidate = get_candidate_by_id(candidate_id) if candidate_id else None
+    transcription_pending = False
+    awaiting_for = session.get("awaiting_transcription_for")
+
+    if candidate and awaiting_for == str(candidate["_id"]):
+        audio_transcription = candidate.get("audio_transcription")
+        if audio_transcription and audio_transcription.get("status") in {"completed", "failed"}:
+            session.pop("awaiting_transcription_for", None)
+        else:
+            transcription_pending = True
+    elif not candidate:
+        session.pop("awaiting_transcription_for", None)
+
+    candidates = get_all_candidates()
+    return render_template(
+        "results.html",
+        candidate=candidate,
+        candidate_count=len(candidates),
+        transcription_pending=transcription_pending
+    )
+
+# ── API: Transcription Status (AJAX polling) ──
+@app.route("/api/transcription_status/<candidate_id>")
+@login_required
+def transcription_status(candidate_id):
+    candidate = get_candidate_by_id(candidate_id)
+    if not candidate:
+        return jsonify({"status": "not_found"}), 404
+    
+    audio = candidate.get("audio_transcription")
+    if not audio:
+        return jsonify({"status": "none"})
+    
+    return jsonify({
+        "status": audio.get("status", "unknown"),
+        "text": audio.get("text", ""),
+        "language": audio.get("language", ""),
+        "error": audio.get("error"),
+    })
+
+# ── API: Candidate Profile (for modal) ──
+@app.route("/api/candidate/<candidate_id>")
+@login_required
+def candidate_profile(candidate_id):
+    candidate = get_candidate_by_id(candidate_id)
+    if not candidate:
+        return jsonify({"error": "not found"}), 404
+    
+    # Don't send the full resume text to keep response small
+    return jsonify({
+        "_id": candidate["_id"],
+        "name": candidate.get("name", "Unknown"),
+        "filename": candidate.get("filename", ""),
+        "resume_url": candidate.get("resume_url", ""),
+        "ai_summary": candidate.get("ai_summary", ""),
+        "skills": candidate.get("skills", []),
+        "education": candidate.get("education", []),
+        "experience": candidate.get("experience", []),
+        "match_score": candidate.get("match_score", 0),
+        "skill_gaps": candidate.get("skill_gaps", {"matched": [], "missing": []}),
+        "job_description": candidate.get("job_description", ""),
+        "audio_transcription": candidate.get("audio_transcription"),
+        "raw_text_preview": candidate.get("raw_text_preview", ""),
+    })
+
+@app.route("/ranking", methods=["GET", "POST"])
+@login_required
+def ranking():
+    candidates = get_all_candidates()
+    job_description = ""
+    ranked = list(candidates)
+
+    if request.method == "POST":
+        job_description = request.form.get("job_description", "").strip()
+        if job_description and candidates:
+            from skill_extractor import extract_skills, SKILLS_LIST
+            from job_matcher import calculate_match_score, find_skill_gaps, rank_candidates
+            from db import candidates_collection
+            from bson.objectid import ObjectId
+            jd_skills = extract_skills(job_description)
+            
+            # Recalculate score and gaps for all candidates and update in DB
+            for candidate in candidates:
+                candidate_id = candidate["_id"]
+                candidate_skills = candidate.get("skills", [])
+                resume_text = candidate.get("resume_text", "")
+                
+                # Calculate new metrics based on new JD
+                new_match_score = calculate_match_score(resume_text, job_description, candidate_skills, jd_skills)
+                new_skill_gaps = find_skill_gaps(candidate_skills, job_description, SKILLS_LIST, jd_skills)
+                
+                # Update DB directly
+                try:
+                    candidates_collection.update_one(
+                        {"_id": ObjectId(candidate_id)},
+                        {"$set": {
+                            "job_description": job_description,
+                            "match_score": new_match_score,
+                            "skill_gaps": new_skill_gaps
+                        }}
+                    )
+                except Exception as e:
+                    logger.error(f"Error updating candidate {candidate_id} during re-ranking: {e}")
+
+            # Re-fetch the updated candidates from the database
+            candidates = get_all_candidates()
+            # Rank the newly fetched candidates
+            ranked = rank_candidates(candidates, job_description, jd_skills)
+
+    return render_template("ranking.html", ranked=ranked, job_description=job_description, candidate_count=len(candidates))
+
+@app.route("/clear")
+@login_required
+def clear():
+    clear_all_candidates()
+    session.pop("last_candidate_id", None)
+    flash("All candidate data cleared from database.", "info")
+    return redirect(url_for("index"))
+
+if __name__ == "__main__":
+    app.run(debug=True, port=5000)

audio_transcriber.py

@@ -0,0 +1,325 @@
+"""
+audio_transcriber.py
+--------------------
+Converts audio files to text using OpenAI's Whisper model.
+Supports both local file paths and URLs.
+Requires ffmpeg to be installed on the system.
+"""
+
+import logging
+import os
+import tempfile
+import threading
+import time
+from urllib.parse import urlparse
+
+import requests
+import whisper
+
+# ── Auto-inject ffmpeg PATH (Windows) ──
+# If ffmpeg is installed at C:\ffmpeg but not in system PATH (common when
+# installed without admin rights), add it to the current process PATH.
+_FFMPEG_COMMON_PATHS = [
+    r"C:\ffmpeg",
+    r"C:\ffmpeg\bin",
+    r"C:\Program Files\ffmpeg\bin",
+    r"C:\Program Files (x86)\ffmpeg\bin",
+]
+for _fp in _FFMPEG_COMMON_PATHS:
+    if os.path.isfile(os.path.join(_fp, "ffmpeg.exe")):
+        if _fp not in os.environ.get("PATH", ""):
+            os.environ["PATH"] = _fp + os.pathsep + os.environ.get("PATH", "")
+            break
+
+# Cache the model so it's only loaded once
+_model = None
+_model_lock = threading.Lock()
+logger = logging.getLogger(__name__)
+
+# Map Content-Type to file extensions
+CONTENT_TYPE_TO_EXT = {
+    "audio/mpeg": ".mp3",
+    "audio/mp3": ".mp3",
+    "audio/wav": ".wav",
+    "audio/x-wav": ".wav",
+    "audio/wave": ".wav",
+    "audio/x-m4a": ".m4a",
+    "audio/mp4": ".m4a",
+    "audio/m4a": ".m4a",
+    "audio/flac": ".flac",
+    "audio/x-flac": ".flac",
+    "audio/ogg": ".ogg",
+    "audio/webm": ".webm",
+    "video/webm": ".webm",
+    "video/mp4": ".mp4",
+    "application/octet-stream": ".mp3",  # fallback for generic binary
+}
+
+
+def _get_model(model_name="base"):
+    """
+    Load and cache the Whisper model.
+    
+    Available models (smallest to largest):
+        tiny, base, small, medium, large
+    
+    Args:
+        model_name (str): Which Whisper model to use.
+    
+    Returns:
+        whisper.Whisper: The loaded model.
+    """
+    global _model
+    if _model is None:
+        with _model_lock:
+            if _model is None:
+                logger.info("Loading Whisper model '%s'...", model_name)
+                _model = whisper.load_model(model_name)
+                logger.info("Whisper model loaded.")
+    return _model
+
+
+def _download_audio_to_tempfile(source_url, timeout=120, max_retries=3):
+    """
+    Download audio from a URL to a temporary file.
+    Includes retry logic, proper extension detection, and User-Agent header.
+    """
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+    }
+
+    last_error = None
+    for attempt in range(1, max_retries + 1):
+        try:
+            logger.info(
+                "Downloading audio (attempt %d/%d): %s", attempt, max_retries, source_url
+            )
+            response = requests.get(
+                source_url, stream=True, timeout=timeout, headers=headers
+            )
+            response.raise_for_status()
+
+            # Determine file extension from Content-Type header
+            content_type = response.headers.get("Content-Type", "").split(";")[0].strip().lower()
+            ext = CONTENT_TYPE_TO_EXT.get(content_type)
+
+            # Fallback: try to extract extension from URL path
+            if not ext:
+                parsed = urlparse(source_url)
+                url_ext = os.path.splitext(parsed.path)[1].lower()
+                if url_ext in {".mp3", ".wav", ".m4a", ".flac", ".ogg", ".webm", ".mp4"}:
+                    ext = url_ext
+                else:
+                    ext = ".mp3"  # safe default for Whisper
+
+            logger.info("Detected content type: %s -> extension: %s", content_type, ext)
+
+            # Write to temp file
+            tmp = tempfile.NamedTemporaryFile(delete=False, suffix=ext)
+            total_bytes = 0
+            try:
+                for chunk in response.iter_content(chunk_size=1024 * 1024):
+                    if chunk:
+                        tmp.write(chunk)
+                        total_bytes += len(chunk)
+            finally:
+                tmp.close()
+
+            logger.info("Downloaded %d bytes to %s", total_bytes, tmp.name)
+
+            if total_bytes == 0:
+                os.unlink(tmp.name)
+                raise RuntimeError("Downloaded file is empty (0 bytes)")
+
+            return tmp.name
+
+        except Exception as e:
+            last_error = e
+            logger.warning("Download attempt %d failed: %s", attempt, e)
+            if attempt < max_retries:
+                wait = 2 ** attempt
+                logger.info("Retrying in %d seconds...", wait)
+                time.sleep(wait)
+
+    raise RuntimeError(
+        f"Failed to download audio after {max_retries} attempts. Last error: {last_error}"
+    )
+
+
+def _check_ffmpeg():
+    """Check if ffmpeg is installed and available in PATH."""
+    import subprocess
+    try:
+        subprocess.run(
+            ["ffmpeg", "-version"],
+            stdout=subprocess.DEVNULL,
+            stderr=subprocess.DEVNULL,
+            timeout=5
+        )
+        return True
+    except (FileNotFoundError, OSError):
+        return False
+
+
+def transcribe_from_local_file(local_path, model_name="base"):
+    """
+    Transcribe audio directly from a local file path.
+    This is the PRIMARY method — saves the audio locally first,
+    then runs Whisper on it.  No network download needed.
+    
+    Args:
+        local_path (str): Absolute path to the audio file on disk.
+        model_name (str): Whisper model size (default: "base").
+    
+    Returns:
+        dict: {
+            "text": str,        # Full transcription
+            "language": str,    # Detected language
+            "success": bool,    # Whether transcription succeeded
+            "error": str|None   # Error message if failed
+        }
+    """
+    try:
+        # --- Check ffmpeg first (WinError 2 prevention) ---
+        if not _check_ffmpeg():
+            msg = (
+                "ffmpeg not found! Whisper needs ffmpeg to decode audio. "
+                "Install it: Windows → 'winget install ffmpeg' then restart your terminal. "
+                "Or download from https://ffmpeg.org/download.html and add to PATH."
+            )
+            logger.error(msg)
+            return {"text": "", "language": "", "success": False, "error": msg}
+
+        if not os.path.exists(local_path):
+            return {
+                "text": "",
+                "language": "",
+                "success": False,
+                "error": f"Local audio file not found: {local_path}",
+            }
+
+        file_size = os.path.getsize(local_path)
+        if file_size == 0:
+            return {
+                "text": "",
+                "language": "",
+                "success": False,
+                "error": "Audio file is empty (0 bytes)",
+            }
+
+        logger.info("Transcribing local file: %s (%d bytes)", local_path, file_size)
+        model = _get_model(model_name)
+        result = model.transcribe(local_path)
+        text = result.get("text", "").strip()
+        language = result.get("language", "unknown")
+        logger.info("Transcription complete. Language: %s, Length: %d chars", language, len(text))
+
+        return {
+            "text": text,
+            "language": language,
+            "success": True,
+            "error": None,
+        }
+
+    except Exception as e:
+        error_msg = str(e)
+        logger.exception("Transcription failed for local file: %s", local_path)
+
+        # WinError 2 = file not found = typically ffmpeg not in PATH
+        if "winerror 2" in error_msg.lower() or "[winerror 2]" in error_msg.lower() or "cannot find the file" in error_msg.lower():
+            error_msg = (
+                "ffmpeg not found in PATH (WinError 2). "
+                "Install it: run 'winget install ffmpeg' in PowerShell as Administrator, then restart. "
+                "Or download from https://ffmpeg.org/download.html"
+            )
+        elif "ffmpeg" in error_msg.lower():
+            error_msg = (
+                "ffmpeg error. Please install ffmpeg and ensure it is in your system PATH. "
+                "Windows: 'winget install ffmpeg'"
+            )
+
+        return {
+            "text": "",
+            "language": "",
+            "success": False,
+            "error": error_msg,
+        }
+
+
+def transcribe_audio(audio_source, model_name="base"):
+    """
+    Transcribe an audio file to text using Whisper.
+    
+    Supported formats: mp3, wav, m4a, flac, ogg, webm
+    
+    Args:
+        audio_source (str): Path or URL to the audio file.
+        model_name (str): Whisper model size (default: "base").
+    
+    Returns:
+        dict: {
+            "text": str,        # Full transcription
+            "language": str,    # Detected language
+            "success": bool,    # Whether transcription succeeded
+            "error": str|None   # Error message if failed
+        }
+    """
+    local_path = None
+    is_url = False
+
+    try:
+        model = _get_model(model_name)
+        is_url = str(audio_source).startswith(("http://", "https://"))
+
+        if is_url:
+            logger.info("Audio source is a URL, downloading: %s", audio_source)
+            local_path = _download_audio_to_tempfile(audio_source)
+        else:
+            local_path = audio_source
+            if not os.path.exists(local_path):
+                return {
+                    "text": "",
+                    "language": "",
+                    "success": False,
+                    "error": f"Local audio file not found: {local_path}",
+                }
+
+        logger.info("Starting Whisper transcription on: %s", local_path)
+        result = model.transcribe(local_path)
+        text = result.get("text", "").strip()
+        language = result.get("language", "unknown")
+        logger.info("Transcription complete. Language: %s, Length: %d chars", language, len(text))
+
+        return {
+            "text": text,
+            "language": language,
+            "success": True,
+            "error": None,
+        }
+
+    except Exception as e:
+        error_msg = str(e)
+        logger.exception("Transcription failed for source: %s", audio_source)
+
+        # Check for common ffmpeg error
+        if "ffmpeg" in error_msg.lower():
+            error_msg = (
+                "ffmpeg is not installed or not found in PATH. "
+                "Please install ffmpeg: https://ffmpeg.org/download.html"
+            )
+
+        return {
+            "text": "",
+            "language": "",
+            "success": False,
+            "error": error_msg,
+        }
+
+    finally:
+        # Clean up temp file only if we downloaded it
+        if is_url and local_path and os.path.exists(local_path):
+            try:
+                os.remove(local_path)
+                logger.info("Cleaned up temp file: %s", local_path)
+            except Exception:
+                pass

db.py

@@ -0,0 +1,143 @@
+"""
+db.py
+-----
+Handles MongoDB connection and operations for HireScope AI.
+"""
+
+import logging
+import os
+from pymongo import MongoClient
+from werkzeug.security import generate_password_hash, check_password_hash
+from datetime import datetime
+from dotenv import load_dotenv
+from bson.objectid import ObjectId
+
+load_dotenv()
+logger = logging.getLogger(__name__)
+
+# Initialize MongoDB Connection
+MONGO_URI = os.getenv("MONGO_URI", "")
+if not MONGO_URI:
+    logger.error("MONGO_URI environment variable is not set!")
+    raise RuntimeError("MONGO_URI environment variable is required. Set it in your .env file.")
+
+client = MongoClient(MONGO_URI)
+
+# Database Name
+db = client.get_database("resume_screener")
+
+# Collections
+users_collection = db.get_collection("users")
+jobs_collection = db.get_collection("jobs")
+candidates_collection = db.get_collection("candidates")
+
+# ── User Operations ──
+
+def create_user(username, email, password, role="recruiter"):
+    if users_collection.find_one({"email": email}):
+        return False, "Email already exists"
+    
+    hashed_password = generate_password_hash(password)
+    user_data = {
+        "username": username,
+        "email": email,
+        "password": hashed_password,
+        "role": role,
+        "created_at": datetime.utcnow()
+    }
+    users_collection.insert_one(user_data)
+    return True, "User created successfully"
+
+def authenticate_user(email, password):
+    user = users_collection.find_one({"email": email})
+    if user and check_password_hash(user["password"], password):
+        # Don't return the password hash in the user object
+        user['_id'] = str(user['_id'])
+        del user['password']
+        return user
+    return None
+
+# ── Candidate Operations ──
+
+def insert_candidate(candidate_data):
+    """
+    Candidate data should include name, filename, text, skills, match_score, etc.
+    """
+    candidate_data["created_at"] = datetime.utcnow()
+    result = candidates_collection.insert_one(candidate_data)
+    return str(result.inserted_id)
+
+def get_all_candidates():
+    candidates = list(candidates_collection.find().sort("match_score", -1))
+    for c in candidates:
+        c['_id'] = str(c['_id'])
+    return candidates
+
+def get_candidate_by_id(candidate_id):
+    try:
+        candidate = candidates_collection.find_one({"_id": ObjectId(candidate_id)})
+        if candidate:
+            candidate['_id'] = str(candidate['_id'])
+        return candidate
+    except Exception:
+        return None
+
+def _candidate_filter(candidate_id):
+    return {"_id": ObjectId(candidate_id)}
+
+
+def set_candidate_audio_processing(candidate_id, audio_url=None):
+    candidates_collection.update_one(
+        _candidate_filter(candidate_id),
+        {
+            "$set": {
+                "audio_transcription": {
+                    "status": "processing",
+                    "text": "",
+                    "language": "",
+                    "error": None,
+                    "audio_url": audio_url,
+                    "updated_at": datetime.utcnow(),
+                }
+            }
+        },
+    )
+
+
+def update_candidate_audio(candidate_id, audio_text, language, audio_url=None):
+    candidates_collection.update_one(
+        _candidate_filter(candidate_id),
+        {
+            "$set": {
+                "audio_transcription": {
+                    "status": "completed",
+                    "text": audio_text,
+                    "language": language,
+                    "error": None,
+                    "audio_url": audio_url,
+                    "updated_at": datetime.utcnow(),
+                }
+            }
+        },
+    )
+
+
+def update_candidate_audio_error(candidate_id, error_msg, audio_url=None):
+    candidates_collection.update_one(
+        _candidate_filter(candidate_id),
+        {
+            "$set": {
+                "audio_transcription": {
+                    "status": "failed",
+                    "text": "",
+                    "language": "",
+                    "error": error_msg,
+                    "audio_url": audio_url,
+                    "updated_at": datetime.utcnow(),
+                }
+            }
+        },
+    )
+
+def clear_all_candidates():
+    candidates_collection.delete_many({})

job_matcher.py

@@ -0,0 +1,122 @@
+"""
+job_matcher.py
+--------------
+Matches resumes to job descriptions using Advanced NLP embeddings (sentence-transformers)
+and performs Exact Skill Overlap measurement. Also provides precise skill gap analysis.
+"""
+import re
+
+from sentence_transformers import SentenceTransformer, util
+
+# Load the model once when the file is imported
+# This will download ~90MB on first run.
+print("Loading sentence-transformers model...")
+model = SentenceTransformer('all-MiniLM-L6-v2')
+
+def calculate_match_score(resume_text, job_description, resume_skills=None, jd_skills=None):
+    """
+    Calculate how well a resume matches a job description using a Hybrid Score:
+    50% Sentence Embeddings (Semantic Match) + 50% Exact Skill Overlap.
+    """
+    if not resume_text.strip() or not job_description.strip():
+        return 0.0
+
+    # 1. Compute Semantic Score (Cosine Similarity)
+    resume_emb = model.encode(resume_text, convert_to_tensor=True)
+    job_emb = model.encode(job_description, convert_to_tensor=True)
+    cosine_scores = util.cos_sim(resume_emb, job_emb)
+    semantic_score = max(0.0, float(cosine_scores[0][0])) * 100
+    
+    # 2. Compute Exact Overlap Score
+    overlap_score = 0.0
+    if jd_skills is not None and resume_skills is not None:
+        jd_skills_lower = [s.lower() for s in jd_skills]
+        resume_skills_lower = [s.lower() for s in resume_skills]
+        if len(jd_skills_lower) > 0:
+            matched_skills = [s for s in jd_skills_lower if s in resume_skills_lower]
+            overlap_score = (len(matched_skills) / len(jd_skills_lower)) * 100
+        
+        # Hybrid Approach: Average of semantic and exact overlap
+        final_score = (semantic_score + overlap_score) / 2
+    else:
+        final_score = semantic_score
+
+    return round(final_score, 1)
+
+
+def find_skill_gaps(resume_skills, job_description, all_skills_list=None, jd_skills=None):
+    """
+    Compare resume skills against skills mentioned in the job description using precise word boundaries.
+    """
+    job_lower = job_description.lower()
+    resume_skills_lower = [s.lower() for s in resume_skills]
+
+    # If jd_skills are explicitly provided, use them
+    if jd_skills is not None:
+        job_skills = jd_skills
+    # Otherwise fallback to regex matching against the all_skills_list
+    elif all_skills_list:
+        job_skills = []
+        for skill in all_skills_list:
+            pattern = r"(?<![a-z0-9])" + re.escape(skill.lower()) + r"(?![a-z0-9])"
+            if re.search(pattern, job_lower):
+                job_skills.append(skill)
+    else:
+        job_skills = []
+
+    matched = [s for s in job_skills if s.lower() in resume_skills_lower]
+    missing = [s for s in job_skills if s.lower() not in resume_skills_lower]
+
+    return {
+        "matched": matched,
+        "missing": missing,
+    }
+
+
+def rank_candidates(candidates, job_description, jd_skills=None):
+    """
+    Rank multiple candidates based on their semantic and exact skill match to a job description.
+    """
+    ranked = []
+    
+    if not job_description.strip():
+        return sorted(candidates, key=lambda x: x.get("match_score", 0), reverse=True)
+
+    # Pre-encode the JD once
+    job_emb = model.encode(job_description, convert_to_tensor=True)
+
+    jd_skills_lower = []
+    if jd_skills:
+        jd_skills_lower = [s.lower() for s in jd_skills]
+
+    for candidate in candidates:
+        resume_emb = model.encode(candidate["resume_text"], convert_to_tensor=True)
+        cosine_scores = util.cos_sim(resume_emb, job_emb)
+        semantic_score = max(0.0, float(cosine_scores[0][0])) * 100
+        
+        # Calculate overlap score
+        overlap_score = 0.0
+        resume_skills = candidate.get("skills", [])
+        resume_skills_lower = [s.lower() for s in resume_skills]
+        
+        if jd_skills_lower:
+            matched_skills = [s for s in jd_skills_lower if s in resume_skills_lower]
+            overlap_score = (len(matched_skills) / len(jd_skills_lower)) * 100
+            final_score = (semantic_score + overlap_score) / 2
+        else:
+            final_score = semantic_score
+            
+        score = round(final_score, 1)
+        
+        ranked.append({
+            "_id": candidate["_id"],
+            "name": candidate["name"],
+            "match_score": score,
+            "skill_count": len(resume_skills),
+            "skills": resume_skills,
+        })
+
+    # Sort by match score descending, then by skill count descending
+    ranked.sort(key=lambda x: (x["match_score"], x["skill_count"]), reverse=True)
+    return ranked
+ 

requirements.txt

@@ -0,0 +1,15 @@
+Flask==3.1.*
+PyPDF2>=3.0
+python-docx>=1.0
+scikit-learn>=1.4
+nltk>=3.9
+openai-whisper>=20231117
+pymongo>=4.6
+sentence-transformers>=2.5
+Werkzeug>=3.0
+Flask-Session>=0.5
+google-generativeai>=0.8
+python-dotenv>=1.0.1
+cloudinary>=1.40.0
+gunicorn>=21.2.0
+requests>=2.31

resume_parser.py

@@ -0,0 +1,192 @@
+"""
+resume_parser.py
+----------------
+Handles extracting text from PDF and DOCX resume files,
+extracting candidate name, phone, email, and
+cleaning the raw text for further processing.
+"""
+
+import re
+import os
+from PyPDF2 import PdfReader
+from docx import Document
+
+
+def extract_text_from_pdf(filepath):
+    """
+    Extract all text from a PDF file.
+    
+    Args:
+        filepath (str): Path to the PDF file.
+    
+    Returns:
+        str: Extracted text from all pages.
+    """
+    text = ""
+    try:
+        reader = PdfReader(filepath)
+        for page in reader.pages:
+            page_text = page.extract_text()
+            if page_text:
+                text += page_text + "\n"
+    except Exception as e:
+        print(f"[ERROR] Failed to read PDF: {e}")
+    return text
+
+
+def extract_text_from_docx(filepath):
+    """
+    Extract all text from a DOCX file, including tables.
+    
+    Args:
+        filepath (str): Path to the DOCX file.
+    
+    Returns:
+        str: Extracted text from all paragraphs and tables.
+    """
+    text = ""
+    try:
+        doc = Document(filepath)
+
+        # Extract paragraphs
+        for para in doc.paragraphs:
+            text += para.text + "\n"
+
+        # Extract text from tables (e.g. skills in tabular format)
+        for table in doc.tables:
+            for row in table.rows:
+                row_text = " | ".join(cell.text.strip() for cell in row.cells if cell.text.strip())
+                if row_text:
+                    text += row_text + "\n"
+    except Exception as e:
+        print(f"[ERROR] Failed to read DOCX: {e}")
+    return text
+
+
+def extract_text(filepath):
+    """
+    Detect file type and extract text accordingly.
+    
+    Args:
+        filepath (str): Path to a PDF or DOCX file.
+    
+    Returns:
+        str: Extracted raw text.
+    
+    Raises:
+        ValueError: If the file format is not supported.
+    """
+    ext = os.path.splitext(filepath)[1].lower()
+
+    if ext == ".pdf":
+        return extract_text_from_pdf(filepath)
+    elif ext == ".docx":
+        return extract_text_from_docx(filepath)
+    else:
+        raise ValueError(f"Unsupported file format: {ext}. Use PDF or DOCX.")
+
+
+def extract_email(raw_text):
+    """
+    Extract email addresses from resume text.
+    
+    Args:
+        raw_text (str): The raw extracted text.
+    
+    Returns:
+        str: First email found, or empty string.
+    """
+    pattern = r'[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}'
+    emails = re.findall(pattern, raw_text)
+    return emails[0] if emails else ""
+
+
+def extract_phone(raw_text):
+    """
+    Extract phone numbers from resume text.
+    Supports Indian (+91), US (+1), and international formats.
+    
+    Args:
+        raw_text (str): The raw extracted text.
+    
+    Returns:
+        str: First phone number found, or empty string.
+    """
+    patterns = [
+        r'(?:\+91[\s-]?)?[6-9]\d{4}[\s-]?\d{5}',     # Indian: +91 98765 43210
+        r'(?:\+1[\s-]?)?\(?\d{3}\)?[\s.-]?\d{3}[\s.-]?\d{4}',  # US: (555) 123-4567
+        r'\+?\d{1,3}[\s.-]?\d{3,4}[\s.-]?\d{3,4}[\s.-]?\d{0,4}',  # International
+    ]
+    for pattern in patterns:
+        phones = re.findall(pattern, raw_text)
+        if phones:
+            # Return the longest match (most likely a real phone number)
+            return max(phones, key=len).strip()
+    return ""
+
+
+def extract_candidate_name(raw_text):
+    """
+    Attempt to extract the candidate's name from the first few lines of the resume.
+    Usually the first non-empty, non-email, non-phone line is the name.
+    
+    Args:
+        raw_text (str): The raw extracted text.
+    
+    Returns:
+        str: Candidate name or empty string.
+    """
+    lines = raw_text.strip().split("\n")
+    for line in lines[:5]:  # Check first 5 lines
+        line = line.strip()
+        if not line:
+            continue
+        # Skip if it's an email
+        if re.search(r'[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}', line):
+            continue
+        # Skip if it's a phone number
+        if re.search(r'[\+]?\d[\d\s\-\(\)]{7,}', line):
+            continue
+        # Skip common headers
+        skip_words = ["resume", "curriculum vitae", "cv", "objective", "summary", "profile"]
+        if line.lower().strip() in skip_words:
+            continue
+        # If line is short and contains mostly letters, it's likely a name
+        if len(line) < 60 and re.match(r'^[A-Za-z\s\.\-]+$', line):
+            return line.title()
+    return ""
+
+
+def clean_text(raw_text):
+    """
+    Clean and normalize extracted text.
+    
+    Steps:
+        1. Convert to lowercase
+        2. Remove URLs
+        3. Remove email addresses
+        4. Remove special characters (keep letters, numbers, spaces, and +, #, -, ., /)
+        5. Collapse multiple spaces into one
+        6. Strip leading/trailing whitespace
+    
+    Args:
+        raw_text (str): The raw extracted text.
+    
+    Returns:
+        str: Cleaned text ready for NLP processing.
+    """
+    text = raw_text.lower()
+
+    # Remove URLs
+    text = re.sub(r"http\S+|www\.\S+", "", text)
+
+    # Remove email addresses
+    text = re.sub(r"\S+@\S+\.\S+", "", text)
+
+    # Remove special characters but keep letters, numbers, spaces, and specific symbols (+, #, -, ., /)
+    text = re.sub(r"[^a-z0-9\s\+\#\-\.\/]", " ", text)
+
+    # Collapse multiple spaces
+    text = re.sub(r"\s+", " ", text)
+
+    return text.strip()

skill_extractor.py

@@ -0,0 +1,499 @@
+"""
+skill_extractor.py (Production v5.0 - BTech All Roles)
+-------------------------------------------------------
+Advanced Resume Skill Extractor:
+- Covers ALL common BTech career roles
+- Regex-based extraction (word-boundary safe)
+- Role-wise organized skill database
+- Smart normalization + inferred skills
+- Clean & deduplicated output
+"""
+
+import re
+
+
+# ─────────────────────────────────────────────
+# 🔹 SKILLS DATABASE — BTech ALL ROLES
+# ─────────────────────────────────────────────
+
+SKILLS_LIST = [
+
+    # ── 1. SOFTWARE DEVELOPMENT ───────────────
+    "python", "java", "javascript", "typescript", "c", "c++", "c#",
+    "ruby", "php", "swift", "kotlin", "go", "rust", "scala", "r", "matlab",
+    "html", "css", "react", "react.js", "next.js", "angular", "vue",
+    "redux", "tailwind css", "framer motion",
+    "node.js", "nodejs", "express", "express.js",
+    "django", "flask", "fastapi", "spring boot",
+    "rest api", "rest apis", "websockets", "graphql",
+    "jwt", "oauth", "authentication", "authorization",
+    "mvc", "state management", "api design",
+    "prisma", "mongoose",
+    "vercel", "netlify",
+    "frontend development", "backend development", "full stack", "web development",
+
+    # ── 2. DATA SCIENCE / ML / AI ─────────────
+    "machine learning", "deep learning", "data science", "data analysis",
+    "tensorflow", "pytorch", "keras", "scikit-learn",
+    "pandas", "numpy", "matplotlib", "seaborn", "scipy",
+    "nlp", "natural language processing", "computer vision",
+    "opencv", "hugging face", "llm", "generative ai", "prompt engineering",
+    "feature engineering", "model deployment", "mlops",
+    "regression", "classification", "clustering", "neural network",
+    "random forest", "xgboost", "time series",
+    "artificial intelligence", "chatgpt", "copilot",
+
+    # ── 3. DATA ENGINEERING / ANALYTICS ───────
+    "sql", "mysql", "postgresql", "mongodb", "redis", "firebase",
+    "oracle", "sqlite", "dynamodb", "cassandra",
+    "power bi", "tableau", "excel", "google sheets",
+    "etl", "data pipeline", "apache spark", "hadoop", "kafka",
+    "airflow", "dbt", "snowflake", "bigquery", "data warehouse",
+    "data visualization", "business intelligence",
+
+    # ── 4. DEVOPS / CLOUD ─────────────────────
+    "aws", "azure", "gcp", "docker", "kubernetes", "jenkins",
+    "terraform", "ansible", "ci/cd", "linux", "bash", "shell scripting",
+    "git", "github", "gitlab", "bitbucket",
+    "nginx", "apache", "load balancing", "microservices",
+    "serverless", "lambda", "cloud computing",
+
+    # ── 5. CYBERSECURITY ──────────────────────
+    "network security", "ethical hacking", "penetration testing",
+    "kali linux", "metasploit", "wireshark", "nmap", "burp suite",
+    "cryptography", "ssl", "tls", "firewall", "ids", "ips",
+    "siem", "soc", "vulnerability assessment", "owasp",
+    "digital forensics", "malware analysis", "incident response",
+    "information security", "cyber security",
+
+    # ── 6. EMBEDDED SYSTEMS / IOT ─────────────
+    "embedded c", "arduino", "raspberry pi", "stm32", "esp32",
+    "rtos", "freertos", "uart", "spi", "i2c", "can bus",
+    "iot", "mqtt", "zigbee", "bluetooth", "wifi module",
+    "pcb design", "kicad", "altium", "proteus", "multisim",
+    "microcontroller", "microprocessor", "fpga", "vhdl", "verilog",
+    "signal processing", "sensor integration",
+
+    # ── 7. VLSI / CHIP DESIGN ─────────────────
+    "vlsi", "verilog", "vhdl", "system verilog",
+    "cadence", "synopsys", "mentor graphics",
+    "rtl design", "synthesis", "sta", "place and route",
+    "digital design", "analog design", "asic", "soc design",
+    "dft", "timing analysis", "power analysis",
+
+    # ── 8. MECHANICAL ENGINEERING ─────────────
+    "autocad", "solidworks", "catia", "ansys", "creo",
+    "fusion 360", "3d printing", "additive manufacturing",
+    "cad", "cam", "cfd", "fea", "finite element analysis",
+    "manufacturing processes", "cnc machining",
+    "thermodynamics", "fluid mechanics", "heat transfer",
+    "robotics", "automation", "plc", "scada", "hmi",
+    "lean manufacturing", "six sigma", "quality control",
+
+    # ── 9. CIVIL ENGINEERING ──────────────────
+    "autocad civil", "staad pro", "etabs", "revit", "primavera",
+    "ms project", "civil 3d",
+    "structural analysis", "structural design", "rcc design",
+    "surveying", "gis", "remote sensing", "arcgis",
+    "construction management", "project planning",
+    "soil mechanics", "geotechnical", "foundation design",
+    "highway design", "transportation engineering",
+    "water supply", "sanitation", "irrigation",
+
+    # ── 10. ELECTRICAL ENGINEERING ────────────
+    "power systems", "power electronics", "circuit design",
+    "matlab simulink", "pspice", "ltspice", "labview",
+    "electric vehicles", "battery management system",
+    "solar energy", "wind energy", "renewable energy",
+    "transformer", "motor drives", "inverter", "rectifier",
+    "control systems", "pid controller",
+    "high voltage", "switchgear", "protection relay",
+    "smart grid", "energy audit",
+
+    # ── 11. ELECTRONICS & COMMUNICATION ───────
+    "signal processing", "dsp", "image processing",
+    "communication systems", "wireless communication",
+    "5g", "lte", "antenna design", "rf design",
+    "hfss", "cst", "ads",
+    "optical fiber", "photonics",
+    "digital electronics", "analog electronics",
+    "oscilloscope", "logic analyzer",
+
+    # ── 12. ROBOTICS / AUTOMATION ─────────────
+    "ros", "ros2", "gazebo", "slam", "path planning",
+    "sensor fusion", "robotic arm", "drone", "uav",
+    "autonomous systems", "control theory", "motion planning",
+    "industrial automation", "plc programming",
+
+    # ── 13. PRODUCT MANAGEMENT (Tech) ─────────
+    "product roadmap", "user stories", "agile", "scrum", "kanban",
+    "jira", "confluence", "notion", "trello",
+    "wireframing", "prototyping", "figma", "balsamiq",
+    "market research", "competitive analysis",
+    "a/b testing", "product analytics", "kpi tracking",
+    "stakeholder management", "go to market",
+
+    # ── 14. UI/UX DESIGN ──────────────────────
+    "figma", "adobe xd", "sketch", "invision",
+    "photoshop", "illustrator", "after effects",
+    "user research", "usability testing",
+    "design thinking", "information architecture",
+    "interaction design", "visual design", "typography",
+    "color theory", "responsive design", "accessibility",
+
+    # ── 15. BUSINESS ANALYST / CONSULTING ─────
+    "requirement gathering", "brd", "frd", "use case",
+    "uml", "flowchart", "process mapping", "gap analysis",
+    "business analysis", "functional testing", "uat",
+
+    # ── 16. TESTING / QA ──────────────────────
+    "manual testing", "automation testing", "selenium",
+    "cypress", "playwright", "jest", "pytest",
+    "test cases", "test plan", "bug tracking",
+    "api testing", "postman", "jmeter", "load testing",
+    "performance testing", "regression testing",
+    "black box testing", "white box testing",
+
+    # ── 17. GAME DEVELOPMENT ──────────────────
+    "unity", "unreal engine", "godot",
+    "game design", "level design", "3d modeling",
+    "blender", "maya", "3ds max",
+    "ar", "vr", "mixed reality", "xr",
+    "physics simulation", "shader programming",
+
+    # ── 18. BLOCKCHAIN ────────────────────────
+    "blockchain", "solidity", "ethereum", "web3.js", "ethers.js",
+    "smart contracts", "nft", "defi", "hyperledger",
+    "cryptocurrency", "metamask", "truffle", "hardhat",
+    "ipfs", "decentralized applications", "dapps",
+
+    # ── 19. SOFT SKILLS ───────────────────────
+    "communication", "leadership", "teamwork", "problem solving",
+    "project management", "critical thinking",
+    "time management", "presentation", "collaboration",
+    "analytical thinking", "attention to detail",
+
+    # ── 20. MOBILE DEVELOPMENT ────────────────
+    "react native", "flutter", "dart", "swiftui", "jetpack compose",
+    "android development", "ios development", "mobile development",
+    "xcode", "android studio", "expo",
+
+    # ── 21. DATA / MODERN TOOLS ───────────────
+    "streamlit", "gradio", "langchain", "llamaindex", "pinecone",
+    "chromadb", "weaviate", "vector database", "rag",
+    "data lake", "lakehouse", "delta lake", "databricks",
+    "looker", "metabase", "superset",
+
+    # ── 22. CLOUD CERTIFICATIONS & SERVICES ───
+    "aws lambda", "s3", "ec2", "ecs", "eks", "cloudfront",
+    "azure devops", "azure functions", "cosmos db",
+    "google cloud functions", "cloud run", "vertex ai",
+    "heroku", "railway", "render", "fly.io",
+
+    # ── 23. API & ARCHITECTURE ────────────────
+    "grpc", "soap", "swagger", "openapi",
+    "event driven", "message queue", "rabbitmq",
+    "design patterns", "solid principles", "clean architecture",
+    "domain driven design", "system design",
+
+    # ── 24. VERSION CONTROL & CI/CD ──────────
+    "github actions", "circleci", "travis ci", "argo cd",
+    "helm", "prometheus", "grafana", "elk stack",
+    "datadog", "new relic", "splunk",
+]
+
+
+# ──────────────────────────���──────────────────
+# 🔥 SKILL NORMALIZATION MAP
+# ─────────────────────────────────────────────
+
+SKILL_MAP = {
+    # Variations
+    "react":      ["react.js"],
+    "node.js":    ["nodejs"],
+    "express":    ["express.js"],
+    "rest api":   ["rest apis"],
+    "html":       ["html5"],
+    "css":        ["css3"],
+
+    # Software Dev
+    "full stack":              ["mern", "fullstack", "mean"],
+    "authentication":          ["jwt", "oauth", "auth"],
+    "frontend development":    ["react", "next.js", "angular", "vue", "html", "css"],
+    "backend development":     ["node.js", "express", "django", "flask", "fastapi", "spring boot"],
+    "version control":         ["git", "github", "gitlab"],
+    "database management":     ["sql", "mysql", "postgresql", "mongodb", "redis", "firebase"],
+
+    # AI / ML
+    "deep learning":           ["tensorflow", "keras", "pytorch"],
+    "data science":            ["pandas", "numpy", "data analysis", "machine learning"],
+    "nlp":                     ["llm", "natural language processing", "text classification", "tokenization", "hugging face"],
+    "machine learning":        ["ml", "scikit-learn", "model training", "regression", "classification", "xgboost"],
+    "artificial intelligence": ["ai", "neural network", "deep learning", "machine learning", "generative ai"],
+    "computer vision":         ["image recognition", "object detection", "opencv", "cnn"],
+    "mlops":                   ["model deployment", "mlflow", "kubeflow"],
+    "seaborn":                 ["matplotlib"],
+    "classification":          ["classify", "classification", "predict", "model training"],
+
+    # DevOps / Cloud
+    "devops":                  ["ci/cd", "docker", "kubernetes", "jenkins", "terraform", "ansible"],
+    "cloud computing":         ["aws", "azure", "gcp", "serverless", "lambda"],
+
+    # Security
+    "cyber security":          ["ethical hacking", "penetration testing", "network security", "owasp"],
+    "ethical hacking":         ["kali linux", "metasploit", "burp suite", "nmap"],
+
+    # Embedded / IoT
+    "iot":                     ["mqtt", "arduino", "raspberry pi", "esp32", "zigbee"],
+    "embedded systems":        ["embedded c", "rtos", "microcontroller", "stm32", "esp32"],
+
+    # VLSI
+    "vlsi":                    ["verilog", "vhdl", "system verilog", "rtl design", "asic"],
+
+    # Mechanical
+    "cad":                     ["autocad", "solidworks", "catia", "creo", "fusion 360"],
+    "simulation":              ["ansys", "fea", "cfd", "matlab simulink"],
+    "automation":              ["plc", "scada", "hmi", "industrial automation"],
+    "lean manufacturing":      ["six sigma", "quality control", "kaizen"],
+
+    # Civil
+    "structural design":       ["staad pro", "etabs", "rcc design"],
+    "gis":                     ["arcgis", "remote sensing", "civil 3d"],
+
+    # Electrical
+    "power electronics":       ["inverter", "rectifier", "motor drives", "battery management system"],
+    "renewable energy":        ["solar energy", "wind energy", "electric vehicles"],
+    "control systems":         ["pid controller", "matlab simulink", "labview"],
+
+    # Robotics
+    "robotics":                ["ros", "ros2", "slam", "path planning", "robotic arm", "drone"],
+
+    # Testing
+    "automation testing":      ["selenium", "cypress", "playwright", "jest", "pytest"],
+    "api testing":             ["postman", "jmeter"],
+
+    # Game Dev
+    "game development":        ["unity", "unreal engine", "godot", "game design"],
+    "3d modeling":             ["blender", "maya", "3ds max"],
+    "vr":                      ["virtual reality", "oculus", "steamvr"],
+    "ar":                      ["augmented reality", "arkit", "arcore"],
+
+    # Blockchain
+    "blockchain":              ["solidity", "ethereum", "smart contracts", "web3.js", "dapps"],
+
+    # Data / Analytics
+    "business intelligence":   ["power bi", "tableau", "data visualization"],
+    "data pipeline":           ["apache spark", "kafka", "airflow", "etl"],
+    "data warehouse":          ["snowflake", "bigquery", "redshift"],
+
+    # Mobile
+    "mobile development":      ["react native", "flutter", "android development", "ios development"],
+    "android development":     ["kotlin", "jetpack compose", "android studio"],
+    "ios development":         ["swift", "swiftui", "xcode"],
+
+    # Modern AI/Data
+    "rag":                     ["langchain", "llamaindex", "vector database", "chromadb", "pinecone"],
+    "generative ai":           ["llm", "chatgpt", "copilot", "prompt engineering", "rag", "langchain"],
+
+    # Product / Design
+    "agile":                   ["scrum", "kanban", "jira", "sprint"],
+    "ui/ux":                   ["figma", "adobe xd", "wireframing", "prototyping", "user research"],
+
+     # Soft skills
+    "collaboration": ["collaborated", "teamwork", "team", "worked with"],
+
+     # Feature engineering
+    "feature engineering": [
+        "data preprocessing",
+        "data cleaning",
+        "feature extraction",
+        "data transformation"
+    ],
+
+    # Teamwork
+    "teamwork": [
+        "team",
+        "collaborated",
+        "community",
+        "worked with",
+        "coordinated"
+    ],
+
+    # Analytical thinking
+    "analytical thinking": [
+        "data analysis",
+        "problem solving",
+        "analysis",
+        "model training"
+    ]
+}
+
+
+# ─────────────────────────────────────────────
+# 🔹 EDUCATION KEYWORDS
+# ─────────────────────────────────────────────
+
+EDUCATION_KEYWORDS = [
+    "bachelor", "master", "phd", "doctorate", "diploma",
+    "b.tech", "m.tech", "b.sc", "m.sc", "b.e", "m.e",
+    "bca", "mca", "bba", "mba", "b.com", "m.com",
+    "university", "college", "institute", "school",
+    "computer science", "information technology",
+    "engineering", "mathematics", "physics", "chemistry",
+    "electronics", "electrical", "mechanical", "civil",
+    "degree", "graduation", "post graduation", "certification",
+    "12th", "10th", "higher secondary", "secondary",
+    "iit", "nit", "iiit", "bits", "vit", "lpu", "amity",
+    "cgpa", "gpa", "percentage", "aggregate",
+    "coursework", "specialization", "minor", "major",
+    "data science", "artificial intelligence",
+    "biotechnology", "biomedical", "chemical engineering",
+    "aerospace", "automobile", "industrial engineering",
+]
+
+
+# ─────────────────────────────────────────────
+# 🔹 EXPERIENCE KEYWORDS
+# ─────────────────────────────────────────────
+
+EXPERIENCE_KEYWORDS = [
+    "experience", "years of experience", "worked at", "working at",
+    "intern", "internship", "fresher", "junior", "senior",
+    "lead", "manager", "director", "team lead",
+    "full time", "part time", "freelance", "contract",
+    "responsibilities", "achievements", "projects",
+    "developed", "implemented", "designed", "managed",
+    "built", "created", "maintained", "optimized",
+    "deployed", "architected", "collaborated", "researched",
+    "analyzed", "tested", "automated", "integrated",
+    "mentored", "supervised", "coordinated", "delivered",
+    "contributed", "spearheaded", "launched", "scaled",
+    "improved", "reduced", "increased", "streamlined",
+    "training", "workshop", "hackathon", "open source",
+    "startup", "company", "organization", "firm",
+    "software engineer", "data analyst", "web developer",
+    "ml engineer", "devops engineer", "full stack developer",
+]
+
+
+# ─────────────────────────────────────────────
+# 🔹 HELPER FUNCTION
+# ─────────────────────────────────────────────
+
+def _match(keyword: str, text: str) -> bool:
+    """Word-boundary safe keyword match."""
+    pattern = r"(?<![a-z0-9])" + re.escape(keyword.lower()) + r"(?![a-z0-9])"
+    return bool(re.search(pattern, text))
+
+
+# ─────────────────────────────────────────────
+# 🔹 SKILL EXTRACTION
+# ─────────────────────────────────────────────
+
+def extract_skills(text: str) -> list[str]:
+    """
+    Extract skills from resume text using regex keyword matching.
+
+    Args:
+        text (str): Cleaned resume text.
+
+    Returns:
+        list[str]: Skills found in the text.
+    """
+    text_lower = text.lower()
+    return [skill for skill in SKILLS_LIST if _match(skill, text_lower)]
+
+
+# ─────────────────────────────────────────────
+# 🔥 NORMALIZATION
+# ─────────────────────────────────────────────
+
+def normalize_skills(skills: list[str], text: str) -> list[str]:
+    """
+    Infer higher-level skills from related keywords found in text.
+    Example: "tensorflow" found → "deep learning" automatically added.
+
+    Args:
+        skills (list[str]): Already extracted raw skills.
+        text (str): Original resume text.
+
+    Returns:
+        list[str]: Expanded, deduplicated, sorted skill list.
+    """
+    normalized = set(skills)
+    text_lower = text.lower()
+
+    for main_skill, related_keywords in SKILL_MAP.items():
+        for keyword in related_keywords:
+            if _match(keyword, text_lower):
+                normalized.add(main_skill)
+                break
+
+    return sorted(normalized)
+
+
+# ────────���────────────────────────────────────
+# 🔹 EDUCATION
+# ─────────────────────────────────────────────
+
+def extract_education(text: str) -> list[str]:
+    """
+    Find education-related keywords in resume text.
+
+    Args:
+        text (str): Cleaned resume text.
+
+    Returns:
+        list[str]: Education keywords found.
+    """
+    text_lower = text.lower()
+    return list(set([kw for kw in EDUCATION_KEYWORDS if _match(kw, text_lower)]))
+
+
+# ─────────────────────────────────────────────
+# 🔹 EXPERIENCE
+# ─────────────────────────────────────────────
+
+def extract_experience(text: str) -> list[str]:
+    """
+    Find experience-related keywords in resume text.
+
+    Args:
+        text (str): Cleaned resume text.
+
+    Returns:
+        list[str]: Experience indicators found.
+    """
+    text_lower = text.lower()
+    return list(set([kw for kw in EXPERIENCE_KEYWORDS if _match(kw, text_lower)]))
+
+
+# ─────────────────────────────────────────────
+# 🔹 MAIN PIPELINE
+# ─────────────────────────────────────────────
+
+def extract_all(text: str) -> dict:
+    """
+    Full extraction pipeline:
+      1. Extract raw skills via regex
+      2. Normalize using SKILL_MAP inference
+      3. Extract education and experience
+
+    Args:
+        text (str): Cleaned resume text.
+
+    Returns:
+        dict: {
+            "skills": [...],
+            "education": [...],
+            "experience": [...]
+        }
+    """
+    raw_skills = extract_skills(text)
+    final_skills = normalize_skills(raw_skills, text)
+
+    return {
+        "skills": final_skills,
+        "education": extract_education(text),
+        "experience": extract_experience(text),
+    } 

static/css/style.css

@@ -0,0 +1,84 @@
+/* HireScope AI — Custom Styles (supplements Tailwind CDN) */
+
+/* Base transitions */
+*, *::before, *::after {
+    transition-property: color, background-color, border-color, box-shadow, transform, opacity;
+    transition-timing-function: cubic-bezier(0.4, 0, 0.2, 1);
+    transition-duration: 150ms;
+}
+
+/* Smooth scroll */
+html {
+    scroll-behavior: smooth;
+}
+
+/* Selection color */
+::selection {
+    background: rgba(99, 102, 241, 0.2);
+    color: #312e81;
+}
+
+/* Focus ring */
+*:focus-visible {
+    outline: 2px solid #6366f1;
+    outline-offset: 2px;
+    border-radius: 8px;
+}
+
+/* Form inputs - remove browser defaults */
+input[type="file"] {
+    cursor: pointer;
+}
+
+input[type="file"]::-webkit-file-upload-button {
+    cursor: pointer;
+}
+
+/* Table row hover transition */
+tbody tr {
+    transition: background-color 0.2s ease;
+}
+
+/* Modal animation */
+#candidate-modal:not(.hidden) #modal-content {
+    animation: modalSlideIn 0.3s ease-out;
+}
+
+@keyframes modalSlideIn {
+    from {
+        opacity: 0;
+        transform: scale(0.95) translateY(10px);
+    }
+    to {
+        opacity: 1;
+        transform: scale(1) translateY(0);
+    }
+}
+
+/* Loading spinner */
+.spinner {
+    border: 2px solid rgba(99, 102, 241, 0.2);
+    border-top: 2px solid #6366f1;
+    border-radius: 50%;
+    width: 24px;
+    height: 24px;
+    animation: spin 0.8s linear infinite;
+}
+
+@keyframes spin {
+    to { transform: rotate(360deg); }
+}
+
+/* Print styles */
+@media print {
+    nav, footer, .btn-primary, .btn-secondary, button {
+        display: none !important;
+    }
+    body {
+        background: white !important;
+    }
+    .glass-card {
+        box-shadow: none !important;
+        border: 1px solid #e2e8f0 !important;
+    }
+}

templates/base.html

@@ -0,0 +1,276 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>HireScope AI — Intelligent Resume Screening</title>
+    <meta name="description" content="AI-powered resume screening system with semantic matching, skill extraction, and audio transcription.">
+
+    <!-- Google Fonts -->
+    <link rel="preconnect" href="https://fonts.googleapis.com">
+    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
+    <link href="https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700;800;900&display=swap" rel="stylesheet">
+
+    <!-- Tailwind CSS CDN -->
+    <script src="https://cdn.tailwindcss.com"></script>
+    <script>
+        tailwind.config = {
+            theme: {
+                extend: {
+                    fontFamily: {
+                        sans: ['Inter', 'system-ui', 'sans-serif'],
+                    },
+                    colors: {
+                        brand: {
+                            50:  '#eef2ff',
+                            100: '#e0e7ff',
+                            200: '#c7d2fe',
+                            300: '#a5b4fc',
+                            400: '#818cf8',
+                            500: '#6366f1',
+                            600: '#4f46e5',
+                            700: '#4338ca',
+                            800: '#3730a3',
+                            900: '#312e81',
+                        },
+                        sky: {
+                            50: '#f0f9ff',
+                            400: '#38bdf8',
+                            500: '#0ea5e9',
+                            600: '#0284c7',
+                        }
+                    }
+                }
+            }
+        }
+    </script>
+    <style>
+        * { font-family: 'Inter', system-ui, sans-serif; }
+
+        body {
+            background: linear-gradient(135deg, #f0f4f8 0%, #e8edf5 50%, #f0f0ff 100%);
+            min-height: 100vh;
+        }
+
+        .glass-card {
+            background: rgba(255, 255, 255, 0.85);
+            backdrop-filter: blur(20px);
+            -webkit-backdrop-filter: blur(20px);
+            border: 1px solid rgba(255, 255, 255, 0.9);
+            box-shadow: 0 4px 24px rgba(0, 0, 0, 0.06), 0 1px 2px rgba(0, 0, 0, 0.04);
+        }
+
+        .glass-card:hover {
+            box-shadow: 0 8px 40px rgba(99, 102, 241, 0.08), 0 2px 8px rgba(0, 0, 0, 0.06);
+        }
+
+        .gradient-text {
+            background: linear-gradient(135deg, #6366f1, #0ea5e9);
+            -webkit-background-clip: text;
+            -webkit-text-fill-color: transparent;
+            background-clip: text;
+        }
+
+        .nav-glass {
+            background: rgba(255, 255, 255, 0.92);
+            backdrop-filter: blur(24px);
+            -webkit-backdrop-filter: blur(24px);
+            border-bottom: 1px solid rgba(226, 232, 240, 0.8);
+        }
+
+        @keyframes fadeInUp {
+            from { opacity: 0; transform: translateY(20px); }
+            to { opacity: 1; transform: translateY(0); }
+        }
+
+        @keyframes slideInRight {
+            from { opacity: 0; transform: translateX(20px); }
+            to { opacity: 1; transform: translateX(0); }
+        }
+
+        @keyframes pulse-soft {
+            0%, 100% { opacity: 1; }
+            50% { opacity: 0.6; }
+        }
+
+        .animate-fade-in-up { animation: fadeInUp 0.5s ease-out forwards; }
+        .animate-fade-in-up-delay { animation: fadeInUp 0.5s ease-out 0.1s forwards; opacity: 0; }
+        .animate-fade-in-up-delay-2 { animation: fadeInUp 0.5s ease-out 0.2s forwards; opacity: 0; }
+        .animate-slide-in { animation: slideInRight 0.4s ease-out forwards; }
+        .animate-pulse-soft { animation: pulse-soft 2s ease-in-out infinite; }
+
+        /* Custom scrollbar */
+        ::-webkit-scrollbar { width: 6px; }
+        ::-webkit-scrollbar-track { background: transparent; }
+        ::-webkit-scrollbar-thumb { background: #cbd5e1; border-radius: 3px; }
+        ::-webkit-scrollbar-thumb:hover { background: #94a3b8; }
+
+        /* Smooth transitions for interactive elements */
+        .btn-primary {
+            background: linear-gradient(135deg, #6366f1, #4f46e5);
+            transition: all 0.3s ease;
+        }
+        .btn-primary:hover {
+            background: linear-gradient(135deg, #818cf8, #6366f1);
+            box-shadow: 0 8px 25px rgba(99, 102, 241, 0.3);
+            transform: translateY(-1px);
+        }
+
+        .btn-secondary {
+            background: linear-gradient(135deg, #0ea5e9, #0284c7);
+            transition: all 0.3s ease;
+        }
+        .btn-secondary:hover {
+            background: linear-gradient(135deg, #38bdf8, #0ea5e9);
+            box-shadow: 0 8px 25px rgba(14, 165, 233, 0.3);
+            transform: translateY(-1px);
+        }
+
+        /* Flash message styles */
+        .flash-success { background: linear-gradient(135deg, #ecfdf5, #d1fae5); border-left: 4px solid #10b981; }
+        .flash-error   { background: linear-gradient(135deg, #fef2f2, #fecaca); border-left: 4px solid #ef4444; }
+        .flash-info    { background: linear-gradient(135deg, #eff6ff, #dbeafe); border-left: 4px solid #3b82f6; }
+        .flash-warning { background: linear-gradient(135deg, #fffbeb, #fef3c7); border-left: 4px solid #f59e0b; }
+
+        /* Mobile menu */
+        .mobile-menu { display: none; }
+        .mobile-menu.open { display: flex; }
+
+        /* Modal backdrop */
+        .modal-backdrop {
+            background: rgba(15, 23, 42, 0.5);
+            backdrop-filter: blur(4px);
+        }
+
+        /* Skill tag animation */
+        .skill-tag {
+            transition: all 0.2s ease;
+        }
+        .skill-tag:hover {
+            transform: translateY(-1px);
+            box-shadow: 0 2px 8px rgba(99, 102, 241, 0.2);
+        }
+    </style>
+</head>
+<body class="min-h-screen flex flex-col font-sans antialiased text-slate-800">
+
+    <!-- Navigation -->
+    <nav class="nav-glass sticky top-0 z-50">
+        <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8">
+            <div class="flex items-center justify-between h-16">
+                <!-- Logo -->
+                <div class="flex items-center gap-3">
+                    <div class="bg-gradient-to-br from-brand-500 to-sky-500 p-2 rounded-xl shadow-lg shadow-brand-200">
+                        <svg class="w-6 h-6 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                            <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z"></path>
+                        </svg>
+                    </div>
+                    <a href="/" class="text-xl font-extrabold gradient-text tracking-tight">HireScope AI</a>
+                </div>
+                
+                <!-- Desktop Navigation -->
+                <div class="hidden md:flex items-center gap-2">
+                    {% if session.get('user_id') %}
+                        <a href="/" class="px-4 py-2 text-slate-600 hover:text-brand-600 hover:bg-brand-50 rounded-lg transition font-medium text-sm">
+                            <span class="flex items-center gap-2">
+                                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16v1a3 3 0 003 3h10a3 3 0 003-3v-1m-4-8l-4-4m0 0L8 8m4-4v12"></path></svg>
+                                Upload
+                            </span>
+                        </a>
+                        <a href="/results" class="px-4 py-2 text-slate-600 hover:text-brand-600 hover:bg-brand-50 rounded-lg transition font-medium text-sm">
+                            <span class="flex items-center gap-2">
+                                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 19v-6a2 2 0 00-2-2H5a2 2 0 00-2 2v6a2 2 0 002 2h2a2 2 0 002-2zm0 0V9a2 2 0 012-2h2a2 2 0 012 2v10m-6 0a2 2 0 002 2h2a2 2 0 002-2m0 0V5a2 2 0 012-2h2a2 2 0 012 2v14a2 2 0 01-2 2h-2a2 2 0 01-2-2z"></path></svg>
+                                Results
+                            </span>
+                        </a>
+                        <a href="/ranking" class="px-4 py-2 text-slate-600 hover:text-brand-600 hover:bg-brand-50 rounded-lg transition font-medium text-sm flex items-center gap-2">
+                            <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 7h8m0 0v8m0-8l-8 8-4-4-6 6"></path></svg>
+                            Leaderboard
+                            {% if candidate_count is defined and candidate_count > 0 %}
+                                <span class="bg-brand-500 text-white text-xs px-2 py-0.5 rounded-full font-semibold">{{ candidate_count }}</span>
+                            {% endif %}
+                        </a>
+
+                        <div class="pl-3 ml-2 border-l border-slate-200 flex items-center gap-3">
+                            <div class="flex items-center gap-2 bg-slate-100 px-3 py-1.5 rounded-lg">
+                                <div class="w-7 h-7 bg-gradient-to-br from-brand-500 to-sky-500 rounded-full flex items-center justify-center text-white text-xs font-bold">
+                                    {{ session.username[0]|upper }}
+                                </div>
+                                <span class="text-sm font-medium text-slate-700">{{ session.username }}</span>
+                            </div>
+                            <a href="/logout" class="text-sm text-red-500 hover:text-red-600 hover:bg-red-50 px-3 py-1.5 rounded-lg transition font-medium">Log out</a>
+                        </div>
+                    {% else %}
+                        <a href="/login" class="px-4 py-2 text-slate-600 hover:text-brand-600 font-medium text-sm rounded-lg hover:bg-brand-50 transition">Sign In</a>
+                        <a href="/register" class="btn-primary text-white px-5 py-2 rounded-lg text-sm font-semibold shadow-md">Sign Up Free</a>
+                    {% endif %}
+                </div>
+
+                <!-- Mobile Hamburger -->
+                <button onclick="document.getElementById('mobile-nav').classList.toggle('open')" class="md:hidden p-2 rounded-lg hover:bg-slate-100 transition">
+                    <svg class="w-6 h-6 text-slate-600" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 6h16M4 12h16M4 18h16"></path></svg>
+                </button>
+            </div>
+
+            <!-- Mobile Menu -->
+            <div id="mobile-nav" class="mobile-menu flex-col gap-2 pb-4 md:hidden">
+                {% if session.get('user_id') %}
+                    <a href="/" class="px-4 py-2 text-slate-600 hover:bg-brand-50 rounded-lg font-medium text-sm">Upload</a>
+                    <a href="/results" class="px-4 py-2 text-slate-600 hover:bg-brand-50 rounded-lg font-medium text-sm">Results</a>
+                    <a href="/ranking" class="px-4 py-2 text-slate-600 hover:bg-brand-50 rounded-lg font-medium text-sm">Leaderboard</a>
+                    <a href="/logout" class="px-4 py-2 text-red-500 hover:bg-red-50 rounded-lg font-medium text-sm">Log out</a>
+                {% else %}
+                    <a href="/login" class="px-4 py-2 text-slate-600 hover:bg-brand-50 rounded-lg font-medium text-sm">Sign In</a>
+                    <a href="/register" class="px-4 py-2 text-brand-600 hover:bg-brand-50 rounded-lg font-medium text-sm">Sign Up</a>
+                {% endif %}
+            </div>
+        </div>
+    </nav>
+
+    <!-- Flash Messages -->
+    {% with messages = get_flashed_messages(with_categories=true) %}
+        {% if messages %}
+            <div class="max-w-5xl mx-auto mt-6 px-4 w-full space-y-3">
+                {% for category, message in messages %}
+                    <div class="flash-{{ category }} px-5 py-4 rounded-xl flex items-center gap-3 animate-slide-in shadow-sm">
+                        {% if category == 'success' %}
+                            <svg class="w-5 h-5 text-emerald-600 flex-shrink-0" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12l2 2 4-4m6 2a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+                            <span class="text-emerald-800 font-medium text-sm">{{ message }}</span>
+                        {% elif category == 'error' %}
+                            <svg class="w-5 h-5 text-red-600 flex-shrink-0" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 8v4m0 4h.01M21 12a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+                            <span class="text-red-800 font-medium text-sm">{{ message }}</span>
+                        {% elif category == 'warning' %}
+                            <svg class="w-5 h-5 text-amber-600 flex-shrink-0" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 9v2m0 4h.01m-6.938 4h13.856c1.54 0 2.502-1.667 1.732-2.5L13.732 4c-.77-.833-1.964-.833-2.732 0L4.082 16.5c-.77.833.192 2.5 1.732 2.5z"></path></svg>
+                            <span class="text-amber-800 font-medium text-sm">{{ message }}</span>
+                        {% else %}
+                            <svg class="w-5 h-5 text-blue-600 flex-shrink-0" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 16h-1v-4h-1m1-4h.01M21 12a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+                            <span class="text-blue-800 font-medium text-sm">{{ message }}</span>
+                        {% endif %}
+                    </div>
+                {% endfor %}
+            </div>
+        {% endif %}
+    {% endwith %}
+
+    <!-- Main Content -->
+    <main class="flex-grow flex flex-col pt-6 pb-16">
+        {% block content %}{% endblock %}
+    </main>
+
+    <!-- Footer -->
+    <footer class="border-t border-slate-200 bg-white/60 backdrop-blur py-6">
+        <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8">
+            <div class="flex flex-col md:flex-row items-center justify-between gap-4">
+                <div class="flex items-center gap-2">
+                    <div class="bg-gradient-to-br from-brand-500 to-sky-500 p-1.5 rounded-lg">
+                        <svg class="w-4 h-4 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z"></path></svg>
+                    </div>
+                    <span class="text-sm font-semibold gradient-text">HireScope AI</span>
+                </div>
+                <p class="text-xs text-slate-400">Built with Semantic AI, Whisper & Sentence-Transformers</p>
+            </div>
+        </div>
+    </footer>
+
+</body>
+</html>

templates/index.html

@@ -0,0 +1,181 @@
+{% extends "base.html" %}
+{% block content %}
+<div class="max-w-6xl mx-auto w-full px-4 sm:px-6 lg:px-8 mt-2">
+    
+    <!-- Hero Section -->
+    <div class="text-center mb-10 animate-fade-in-up">
+        <h1 class="text-4xl md:text-5xl font-extrabold text-slate-800 mb-4 tracking-tight leading-tight">
+            Screen Resumes with<br>
+            <span class="gradient-text">Semantic AI Intelligence</span>
+        </h1>
+        <p class="text-lg text-slate-500 max-w-2xl mx-auto leading-relaxed">Upload resumes and match them against job descriptions using advanced AI embeddings. Get instant skill extraction, scoring, and candidate ranking.</p>
+    </div>
+
+    <!-- Stats Cards -->
+    <div class="grid grid-cols-1 sm:grid-cols-3 gap-4 mb-10 animate-fade-in-up-delay">
+        <div class="glass-card rounded-2xl p-5 flex items-center gap-4">
+            <div class="w-12 h-12 bg-gradient-to-br from-brand-100 to-brand-200 rounded-xl flex items-center justify-center flex-shrink-0">
+                <svg class="w-6 h-6 text-brand-600" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M17 20h5v-2a3 3 0 00-5.356-1.857M17 20H7m10 0v-2c0-.656-.126-1.283-.356-1.857M7 20H2v-2a3 3 0 015.356-1.857M7 20v-2c0-.656.126-1.283.356-1.857m0 0a5.002 5.002 0 019.288 0M15 7a3 3 0 11-6 0 3 3 0 016 0zm6 3a2 2 0 11-4 0 2 2 0 014 0zM7 10a2 2 0 11-4 0 2 2 0 014 0z"></path></svg>
+            </div>
+            <div>
+                <p class="text-2xl font-extrabold text-slate-800">{{ candidate_count }}</p>
+                <p class="text-xs text-slate-500 font-medium uppercase tracking-wide">Candidates</p>
+            </div>
+        </div>
+        <div class="glass-card rounded-2xl p-5 flex items-center gap-4">
+            <div class="w-12 h-12 bg-gradient-to-br from-emerald-100 to-emerald-200 rounded-xl flex items-center justify-center flex-shrink-0">
+                <svg class="w-6 h-6 text-emerald-600" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 19v-6a2 2 0 00-2-2H5a2 2 0 00-2 2v6a2 2 0 002 2h2a2 2 0 002-2zm0 0V9a2 2 0 012-2h2a2 2 0 012 2v10m-6 0a2 2 0 002 2h2a2 2 0 002-2m0 0V5a2 2 0 012-2h2a2 2 0 012 2v14a2 2 0 01-2 2h-2a2 2 0 01-2-2z"></path></svg>
+            </div>
+            <div>
+                <p class="text-2xl font-extrabold text-slate-800">{{ avg_score }}%</p>
+                <p class="text-xs text-slate-500 font-medium uppercase tracking-wide">Avg. Score</p>
+            </div>
+        </div>
+        <div class="glass-card rounded-2xl p-5 flex items-center gap-4">
+            <div class="w-12 h-12 bg-gradient-to-br from-violet-100 to-violet-200 rounded-xl flex items-center justify-center flex-shrink-0">
+                <svg class="w-6 h-6 text-violet-600" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19.428 15.428a2 2 0 00-1.022-.547l-2.387-.477a6 6 0 00-3.86.517l-.318.158a6 6 0 01-3.86.517L6.05 15.21a2 2 0 00-1.806.547M8 4h8l-1 1v5.172a2 2 0 00.586 1.414l5 5c1.26 1.26.367 3.414-1.415 3.414H4.828c-1.782 0-2.674-2.154-1.414-3.414l5-5A2 2 0 009 10.172V5L8 4z"></path></svg>
+            </div>
+            <div>
+                <p class="text-2xl font-extrabold text-slate-800">AI</p>
+                <p class="text-xs text-slate-500 font-medium uppercase tracking-wide">Embedding Match</p>
+            </div>
+        </div>
+    </div>
+
+    <!-- Upload Section -->
+    <div class="max-w-3xl mx-auto animate-fade-in-up-delay-2">
+        <div class="glass-card p-8 rounded-2xl relative overflow-hidden">
+            <!-- Decorative gradient blob -->
+            <div class="absolute -top-10 -right-10 w-40 h-40 bg-gradient-to-br from-brand-200/40 to-sky-200/40 rounded-full blur-3xl pointer-events-none"></div>
+            <div class="absolute -bottom-10 -left-10 w-32 h-32 bg-gradient-to-br from-violet-200/30 to-brand-200/30 rounded-full blur-3xl pointer-events-none"></div>
+
+            <div class="relative">
+                <h2 class="text-xl font-bold text-slate-800 mb-1 flex items-center gap-3">
+                    <div class="w-10 h-10 bg-gradient-to-br from-brand-500 to-brand-600 rounded-xl flex items-center justify-center shadow-md shadow-brand-200">
+                        <svg class="w-5 h-5 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z"></path></svg>
+                    </div>
+                    Upload & Analyze Resume
+                </h2>
+                <p class="text-slate-500 text-sm mb-6 ml-13 pl-13">Upload a PDF/DOCX resume and optionally provide a job description for AI-powered matching.</p>
+
+                <form action="/upload" method="POST" enctype="multipart/form-data" class="space-y-5">
+                    <!-- File Upload Area -->
+                    <div>
+                        <label class="block text-sm font-semibold text-slate-700 mb-2">Resume File</label>
+                        <div id="drop-zone" class="w-full flex justify-center px-6 pt-6 pb-6 border-2 border-slate-200 border-dashed rounded-2xl hover:border-brand-400 hover:bg-brand-50/30 transition-all cursor-pointer group">
+                            <div class="space-y-2 text-center">
+                                <div class="mx-auto w-14 h-14 bg-brand-50 rounded-2xl flex items-center justify-center group-hover:bg-brand-100 transition">
+                                    <svg class="w-7 h-7 text-brand-400 group-hover:text-brand-500 transition" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                                        <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M7 16a4 4 0 01-.88-7.903A5 5 0 1115.9 6L16 6a5 5 0 011 9.9M15 13l-3-3m0 0l-3 3m3-3v12"></path>
+                                    </svg>
+                                </div>
+                                <div class="flex text-sm text-slate-500 justify-center">
+                                    <label for="resume-upload" class="relative cursor-pointer rounded-md font-semibold text-brand-600 hover:text-brand-500 transition">
+                                        <span>Choose file</span>
+                                        <input id="resume-upload" name="resume" type="file" class="sr-only" required accept=".pdf,.docx">
+                                    </label>
+                                    <p class="pl-1">or drag & drop</p>
+                                </div>
+                                <p class="text-xs text-slate-400 font-medium" id="resume-filename">PDF or DOCX up to 10MB</p>
+                            </div>
+                        </div>
+                    </div>
+
+                    <!-- Job Description -->
+                    <div>
+                        <label for="job_description" class="block text-sm font-semibold text-slate-700 mb-2">
+                            Job Description 
+                            <span class="text-slate-400 font-normal">(Optional)</span>
+                        </label>
+                        <textarea name="job_description" id="job_description" rows="4" 
+                            class="w-full bg-slate-50 border border-slate-200 rounded-xl p-4 text-slate-800 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition resize-none placeholder-slate-400 text-sm"
+                            placeholder="Paste the job requirements here to generate a semantic match score..."></textarea>
+                    </div>
+                    
+                    <button type="submit" id="analyze-btn" class="w-full btn-primary text-white font-semibold py-3.5 px-4 rounded-xl shadow-lg text-sm flex justify-center items-center gap-2">
+                        <svg class="w-5 h-5" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 10V3L4 14h7v7l9-11h-7z"></path></svg>
+                        Analyze with AI
+                    </button>
+                </form>
+            </div>
+        </div>
+    </div>
+
+    <!-- Recent Candidates -->
+    {% if recent_candidates %}
+    <div class="max-w-3xl mx-auto mt-10 animate-fade-in-up-delay-2">
+        <h3 class="text-lg font-bold text-slate-700 mb-4 flex items-center gap-2">
+            <svg class="w-5 h-5 text-slate-400" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 8v4l3 3m6-3a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+            Recent Candidates
+        </h3>
+        <div class="space-y-3">
+            {% for c in recent_candidates %}
+            <div class="glass-card rounded-xl p-4 flex items-center justify-between group hover:shadow-md transition">
+                <div class="flex items-center gap-3">
+                    <div class="w-10 h-10 bg-gradient-to-br from-brand-100 to-sky-100 rounded-xl flex items-center justify-center text-brand-600 font-bold text-sm">
+                        {{ c.name[0]|upper }}
+                    </div>
+                    <div>
+                        <p class="font-semibold text-slate-700 text-sm">{{ c.name }}</p>
+                        <p class="text-xs text-slate-400">{{ c.skills|length }} skills detected</p>
+                    </div>
+                </div>
+                <div class="flex items-center gap-3">
+                    <span class="text-sm font-bold {% if c.match_score >= 70 %}text-emerald-600{% elif c.match_score >= 40 %}text-amber-600{% else %}text-slate-400{% endif %}">
+                        {{ c.match_score }}%
+                    </span>
+                    <div class="w-16 h-1.5 bg-slate-100 rounded-full overflow-hidden">
+                        <div class="h-full rounded-full {% if c.match_score >= 70 %}bg-emerald-500{% elif c.match_score >= 40 %}bg-amber-400{% else %}bg-slate-300{% endif %}" style="width: {{ c.match_score }}%"></div>
+                    </div>
+                </div>
+            </div>
+            {% endfor %}
+        </div>
+    </div>
+    {% endif %}
+</div>
+
+<script>
+    // File name display
+    document.getElementById('resume-upload').addEventListener('change', function(e) {
+        const name = e.target.files[0]?.name;
+        const el = document.getElementById('resume-filename');
+        if (name) {
+            el.textContent = '📄 ' + name;
+            el.classList.add('text-brand-600', 'font-semibold');
+            el.classList.remove('text-slate-400');
+        }
+    });
+
+    // Drag and drop
+    const dropZone = document.getElementById('drop-zone');
+    const fileInput = document.getElementById('resume-upload');
+
+    ['dragenter', 'dragover'].forEach(evt => {
+        dropZone.addEventListener(evt, function(e) {
+            e.preventDefault();
+            dropZone.classList.add('border-brand-400', 'bg-brand-50/50');
+        });
+    });
+
+    ['dragleave', 'drop'].forEach(evt => {
+        dropZone.addEventListener(evt, function(e) {
+            e.preventDefault();
+            dropZone.classList.remove('border-brand-400', 'bg-brand-50/50');
+        });
+    });
+
+    dropZone.addEventListener('drop', function(e) {
+        e.preventDefault();
+        const files = e.dataTransfer.files;
+        if (files.length) {
+            fileInput.files = files;
+            fileInput.dispatchEvent(new Event('change'));
+        }
+    });
+
+    dropZone.addEventListener('click', function() {
+        fileInput.click();
+    });
+</script>
+{% endblock %}

templates/login.html

@@ -0,0 +1,48 @@
+{% extends "base.html" %}
+{% block content %}
+<div class="flex-grow flex items-center justify-center px-4">
+    <div class="w-full max-w-md animate-fade-in-up">
+        <!-- Brand Header -->
+        <div class="text-center mb-8">
+            <div class="inline-flex items-center justify-center w-16 h-16 bg-gradient-to-br from-brand-500 to-sky-500 rounded-2xl shadow-lg shadow-brand-200 mb-4">
+                <svg class="w-8 h-8 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z"></path>
+                </svg>
+            </div>
+            <h1 class="text-2xl font-extrabold text-slate-800 mb-1">Welcome back</h1>
+            <p class="text-slate-500 text-sm">Sign in to your HireScope AI account</p>
+        </div>
+
+        <!-- Login Card -->
+        <div class="glass-card p-8 rounded-2xl">
+            <form method="POST" action="/login" class="space-y-5">
+                <div>
+                    <label for="email" class="block text-sm font-semibold text-slate-700 mb-1.5">Email Address</label>
+                    <input type="email" name="email" id="email" required
+                        class="w-full bg-slate-50 border border-slate-200 text-slate-800 rounded-xl px-4 py-3 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition placeholder-slate-400"
+                        placeholder="recruiter@company.com">
+                </div>
+                
+                <div>
+                    <label for="password" class="block text-sm font-semibold text-slate-700 mb-1.5">Password</label>
+                    <input type="password" name="password" id="password" required
+                        class="w-full bg-slate-50 border border-slate-200 text-slate-800 rounded-xl px-4 py-3 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition placeholder-slate-400"
+                        placeholder="••••••••">
+                </div>
+                
+                <button type="submit" class="w-full btn-primary text-white font-semibold py-3 rounded-xl shadow-lg text-sm flex items-center justify-center gap-2">
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M11 16l-4-4m0 0l4-4m-4 4h14m-5 4v1a3 3 0 01-3 3H6a3 3 0 01-3-3V7a3 3 0 013-3h7a3 3 0 013 3v1"></path></svg>
+                    Sign In
+                </button>
+            </form>
+        </div>
+
+        <!-- Footer Link -->
+        <div class="mt-6 text-center">
+            <p class="text-slate-500 text-sm">Don't have an account? 
+                <a href="/register" class="text-brand-600 hover:text-brand-700 font-semibold transition">Create one free</a>
+            </p>
+        </div>
+    </div>
+</div>
+{% endblock %}

templates/ranking.html

@@ -0,0 +1,294 @@
+{% extends "base.html" %}
+{% block content %}
+<div class="max-w-7xl mx-auto w-full px-4 sm:px-6 lg:px-8 mt-2 animate-fade-in-up">
+
+    <!-- Header -->
+    <div class="flex flex-col sm:flex-row justify-between items-start sm:items-end mb-8 gap-4">
+        <div>
+            <h1 class="text-3xl font-extrabold text-slate-800 flex items-center gap-3">
+                <div class="w-10 h-10 bg-gradient-to-br from-amber-400 to-orange-500 rounded-xl flex items-center justify-center shadow-md shadow-amber-200">
+                    <svg class="w-5 h-5 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M5 3v4M3 5h4M6 17v4m-2-2h4m5-16l2.286 6.857L21 12l-5.714 2.143L13 21l-2.286-6.857L5 12l5.714-2.143L13 3z"></path></svg>
+                </div>
+                Candidate Leaderboard
+            </h1>
+            <p class="text-slate-500 mt-1 text-sm ml-13">{{ candidate_count }} candidates processed — click any row to view full profile</p>
+        </div>
+        <div class="flex gap-3">
+            <a href="/" class="px-4 py-2.5 bg-white hover:bg-slate-50 text-slate-700 rounded-xl transition border border-slate-200 shadow-sm font-medium text-sm flex items-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 4v16m8-8H4"></path></svg>
+                Add Candidate
+            </a>
+            <a href="/clear" class="px-4 py-2.5 bg-red-50 hover:bg-red-100 text-red-600 rounded-xl transition border border-red-200 shadow-sm font-medium text-sm flex items-center gap-2" onclick="return confirm('⚠️ Are you sure? This will permanently delete ALL candidates from the database.')">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 7l-.867 12.142A2 2 0 0116.138 21H7.862a2 2 0 01-1.995-1.858L5 7m5 4v6m4-6v6m1-10V4a1 1 0 00-1-1h-4a1 1 0 00-1 1v3M4 7h16"></path></svg>
+                Clear All
+            </a>
+        </div>
+    </div>
+
+    <!-- Re-Rank Panel -->
+    <div class="glass-card p-5 rounded-2xl mb-8 animate-fade-in-up-delay">
+        <form method="POST" action="/ranking" class="flex flex-col md:flex-row gap-4 items-end">
+            <div class="flex-grow w-full">
+                <label for="job_description" class="block text-xs font-bold text-slate-500 uppercase tracking-wider mb-2">Re-Rank via Semantic Similarity</label>
+                <input type="text" name="job_description" id="job_description" 
+                    class="w-full bg-slate-50 border border-slate-200 rounded-xl p-3.5 text-slate-800 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition text-sm placeholder-slate-400"
+                    placeholder="Enter keywords or a full job description to re-rank all candidates..." 
+                    value="{{ job_description }}">
+            </div>
+            <button type="submit" class="w-full md:w-auto whitespace-nowrap btn-primary text-white font-semibold py-3.5 px-6 rounded-xl shadow-md text-sm flex items-center justify-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 4v5h.582m15.356 2A8.001 8.001 0 004.582 9m0 0H9m11 11v-5h-.581m0 0a8.003 8.003 0 01-15.357-2m15.357 2H15"></path></svg>
+                Recalculate
+            </button>
+        </form>
+    </div>
+
+    {% if not ranked %}
+        <!-- Empty State -->
+        <div class="glass-card p-16 text-center rounded-2xl">
+            <div class="w-20 h-20 bg-slate-100 rounded-2xl flex items-center justify-center mx-auto mb-6">
+                <svg class="w-10 h-10 text-slate-300" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M17 20h5v-2a3 3 0 00-5.356-1.857M17 20H7m10 0v-2c0-.656-.126-1.283-.356-1.857M7 20H2v-2a3 3 0 015.356-1.857M7 20v-2c0-.656.126-1.283.356-1.857m0 0a5.002 5.002 0 019.288 0M15 7a3 3 0 11-6 0 3 3 0 016 0z"></path></svg>
+            </div>
+            <h3 class="text-xl font-bold text-slate-700 mb-2">No candidates yet</h3>
+            <p class="text-slate-400 mb-6">Upload resumes to start building your leaderboard</p>
+            <a href="/" class="btn-primary text-white px-6 py-3 rounded-xl shadow-md font-semibold text-sm inline-flex items-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 4v16m8-8H4"></path></svg>
+                Upload Resume
+            </a>
+        </div>
+    {% else %}
+        <!-- Leaderboard Table -->
+        <div class="glass-card rounded-2xl overflow-hidden animate-fade-in-up-delay">
+            <div class="overflow-x-auto">
+                <table class="min-w-full">
+                    <thead>
+                        <tr class="bg-gradient-to-r from-slate-50 to-slate-100 border-b border-slate-200">
+                            <th class="px-6 py-4 text-left text-xs font-bold text-slate-500 uppercase tracking-wider">Rank</th>
+                            <th class="px-6 py-4 text-left text-xs font-bold text-slate-500 uppercase tracking-wider">Candidate</th>
+                            <th class="px-6 py-4 text-left text-xs font-bold text-slate-500 uppercase tracking-wider">Match Score</th>
+                            <th class="px-6 py-4 text-left text-xs font-bold text-slate-500 uppercase tracking-wider">Skills</th>
+                            <th class="px-6 py-4 text-left text-xs font-bold text-slate-500 uppercase tracking-wider">Action</th>
+                        </tr>
+                    </thead>
+                    <tbody class="divide-y divide-slate-100">
+                        {% for candidate in ranked %}
+                        <tr class="hover:bg-brand-50/40 transition-colors group cursor-pointer" onclick="openCandidateModal('{{ candidate._id }}')">
+                            <!-- Rank -->
+                            <td class="px-6 py-4 whitespace-nowrap">
+                                {% if loop.index == 1 %}
+                                    <span class="inline-flex items-center justify-center w-9 h-9 rounded-xl bg-gradient-to-br from-amber-300 to-amber-500 text-white font-bold text-sm shadow-md shadow-amber-200">🥇</span>
+                                {% elif loop.index == 2 %}
+                                    <span class="inline-flex items-center justify-center w-9 h-9 rounded-xl bg-gradient-to-br from-slate-300 to-slate-400 text-white font-bold text-sm shadow-md shadow-slate-200">🥈</span>
+                                {% elif loop.index == 3 %}
+                                    <span class="inline-flex items-center justify-center w-9 h-9 rounded-xl bg-gradient-to-br from-amber-600 to-amber-700 text-white font-bold text-sm shadow-md shadow-amber-200">🥉</span>
+                                {% else %}
+                                    <span class="inline-flex items-center justify-center w-9 h-9 rounded-xl bg-slate-100 text-slate-500 font-bold text-sm">{{ loop.index }}</span>
+                                {% endif %}
+                            </td>
+                            <!-- Candidate Name -->
+                            <td class="px-6 py-4 whitespace-nowrap">
+                                <div class="flex items-center gap-3">
+                                    <div class="w-10 h-10 bg-gradient-to-br from-brand-400 to-sky-400 rounded-xl flex items-center justify-center text-white font-bold text-sm shadow-sm flex-shrink-0">
+                                        {{ candidate.name[0]|upper }}
+                                    </div>
+                                    <div>
+                                        <div class="font-semibold text-slate-800 text-sm">{{ candidate.name }}</div>
+                                        <div class="text-xs text-slate-400">ID: {{ candidate._id[-8:] }}</div>
+                                    </div>
+                                </div>
+                            </td>
+                            <!-- Score -->
+                            <td class="px-6 py-4 whitespace-nowrap">
+                                <div class="flex items-center gap-3">
+                                    <span class="text-lg font-extrabold {% if candidate.match_score >= 70 %}text-emerald-600{% elif candidate.match_score >= 40 %}text-amber-600{% else %}text-slate-500{% endif %}">
+                                        {{ candidate.match_score }}%
+                                    </span>
+                                    <div class="w-20 h-2 bg-slate-100 rounded-full overflow-hidden">
+                                        <div class="h-full rounded-full transition-all duration-500 {% if candidate.match_score >= 70 %}bg-gradient-to-r from-emerald-400 to-emerald-500{% elif candidate.match_score >= 40 %}bg-gradient-to-r from-amber-400 to-amber-500{% else %}bg-gradient-to-r from-slate-300 to-slate-400{% endif %}" style="width: {{ candidate.match_score }}%"></div>
+                                    </div>
+                                </div>
+                            </td>
+                            <!-- Skills -->
+                            <td class="px-6 py-4">
+                                <div class="flex flex-wrap gap-1.5 max-w-xs">
+                                    {% for skill in candidate.skills[:4] %}
+                                        <span class="px-2 py-0.5 bg-brand-50 text-brand-600 border border-brand-100 rounded-lg text-xs font-medium">{{ skill }}</span>
+                                    {% endfor %}
+                                    {% if candidate.skills|length > 4 %}
+                                        <span class="px-2 py-0.5 bg-slate-100 text-slate-500 rounded-lg text-xs font-medium">+{{ candidate.skills|length - 4 }} more</span>
+                                    {% endif %}
+                                </div>
+                            </td>
+                            <!-- Action -->
+                            <td class="px-6 py-4 whitespace-nowrap">
+                                <button onclick="event.stopPropagation(); openCandidateModal('{{ candidate._id }}')" class="text-brand-600 hover:text-brand-700 font-semibold text-xs flex items-center gap-1 group-hover:underline">
+                                    View Profile
+                                    <svg class="w-3.5 h-3.5" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 5l7 7-7 7"></path></svg>
+                                </button>
+                            </td>
+                        </tr>
+                        {% endfor %}
+                    </tbody>
+                </table>
+            </div>
+        </div>
+    {% endif %}
+</div>
+
+<!-- Candidate Profile Modal -->
+<div id="candidate-modal" class="fixed inset-0 z-[100] hidden">
+    <div class="modal-backdrop absolute inset-0" onclick="closeModal()"></div>
+    <div class="relative flex items-center justify-center min-h-screen p-4">
+        <div id="modal-content" class="relative bg-white rounded-2xl shadow-2xl w-full max-w-2xl max-h-[85vh] overflow-y-auto p-0 animate-fade-in-up" style="animation-duration: 0.3s;">
+            <!-- Modal will be populated by JS -->
+            <div id="modal-body" class="p-8">
+                <div class="flex items-center justify-center py-12">
+                    <div class="w-8 h-8 border-2 border-brand-500 border-t-transparent rounded-full animate-spin"></div>
+                </div>
+            </div>
+        </div>
+    </div>
+</div>
+
+<script>
+function openCandidateModal(candidateId) {
+    const modal = document.getElementById('candidate-modal');
+    const body = document.getElementById('modal-body');
+    modal.classList.remove('hidden');
+    document.body.style.overflow = 'hidden';
+
+    // Show loading
+    body.innerHTML = `<div class="flex items-center justify-center py-16"><div class="w-8 h-8 border-2 border-indigo-500 border-t-transparent rounded-full animate-spin"></div></div>`;
+
+    fetch(`/api/candidate/${candidateId}`)
+        .then(r => r.json())
+        .then(data => {
+            if (data.error) {
+                body.innerHTML = `<p class="text-red-500 text-center py-12">Candidate not found</p>`;
+                return;
+            }
+
+            const scoreColor = data.match_score >= 70 ? 'emerald' : data.match_score >= 40 ? 'amber' : 'slate';
+
+            let skillsHtml = data.skills.map(s => 
+                `<span class="px-2.5 py-1 bg-indigo-50 text-indigo-700 border border-indigo-100 rounded-lg text-xs font-medium">${s}</span>`
+            ).join('') || '<span class="text-slate-400 text-sm italic">None detected</span>';
+
+            let eduHtml = data.education.map(e => 
+                `<span class="px-2.5 py-1 bg-violet-50 text-violet-700 border border-violet-100 rounded-lg text-xs font-medium">${e}</span>`
+            ).join('') || '<span class="text-slate-400 text-sm italic">None detected</span>';
+
+            let expHtml = data.experience.map(e => 
+                `<span class="px-2.5 py-1 bg-teal-50 text-teal-700 border border-teal-100 rounded-lg text-xs font-medium">${e}</span>`
+            ).join('') || '<span class="text-slate-400 text-sm italic">None detected</span>';
+
+            let matchedHtml = (data.skill_gaps.matched || []).map(s => 
+                `<span class="px-2.5 py-1 bg-emerald-50 text-emerald-700 border border-emerald-100 rounded-lg text-xs font-medium">✓ ${s}</span>`
+            ).join('') || '<span class="text-slate-400 text-xs italic">No JD provided</span>';
+
+            let missingHtml = (data.skill_gaps.missing || []).map(s => 
+                `<span class="px-2.5 py-1 bg-red-50 text-red-600 border border-red-100 rounded-lg text-xs font-medium line-through">${s}</span>`
+            ).join('') || '<span class="text-slate-400 text-xs italic">No gaps</span>';
+
+            let audioHtml = '';
+            if (data.audio_transcription && data.audio_transcription.text) {
+                audioHtml = `
+                    <div class="mt-6 bg-gradient-to-r from-emerald-50 to-teal-50 p-4 rounded-xl border border-emerald-100">
+                        <h4 class="text-xs font-bold text-emerald-700 uppercase mb-2 flex items-center gap-1.5">
+                            <svg class="w-3.5 h-3.5" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 11a7 7 0 01-7 7m0 0a7 7 0 01-7-7m7 7v4m0 0H8m4 0h4m-4-8a3 3 0 01-3-3V5a3 3 0 116 0v6a3 3 0 01-3 3z"></path></svg>
+                            Audio Transcription (${(data.audio_transcription.language || '').toUpperCase()})
+                        </h4>
+                        <p class="text-slate-700 text-sm italic leading-relaxed">"${data.audio_transcription.text}"</p>
+                    </div>`;
+            }
+
+            let resumeLinkHtml = '';
+
+            body.innerHTML = `
+                <!-- Header -->
+                <div class="flex items-center justify-between mb-6">
+                    <div class="flex items-center gap-4">
+                        <div class="w-14 h-14 bg-gradient-to-br from-indigo-500 to-sky-500 rounded-2xl flex items-center justify-center text-white text-xl font-bold shadow-lg shadow-indigo-200">
+                            ${data.name[0].toUpperCase()}
+                        </div>
+                        <div>
+                            <h2 class="text-xl font-bold text-slate-800">${data.name}</h2>
+                            <p class="text-xs text-slate-400">ID: ${data._id} • ${data.skills.length} skills</p>
+                            ${resumeLinkHtml}
+                        </div>
+                    </div>
+                    <button onclick="closeModal()" class="w-8 h-8 bg-slate-100 hover:bg-slate-200 rounded-lg flex items-center justify-center transition">
+                        <svg class="w-4 h-4 text-slate-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M6 18L18 6M6 6l12 12"></path></svg>
+                    </button>
+                </div>
+
+                ${data.ai_summary ? `
+                <!-- Google Gen AI Summary -->
+                <div class="bg-gradient-to-r from-blue-50 to-indigo-50 p-4 rounded-xl border border-blue-100 mb-6">
+                    <div class="text-xs text-blue-600 uppercase font-bold mb-2 flex items-center gap-1.5">
+                        <svg class="w-4 h-4 text-blue-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 10V3L4 14h7v7l9-11h-7z"></path></svg>
+                        AI Profile Summary
+                    </div>
+                    <p class="text-slate-700 text-sm italic leading-relaxed">${data.ai_summary}</p>
+                </div>
+                ` : ''}
+
+                <!-- Score -->
+                <div class="flex items-center gap-4 mb-6 p-4 bg-gradient-to-r from-${scoreColor}-50 to-transparent rounded-xl border border-${scoreColor}-100">
+                    <div class="text-3xl font-extrabold text-${scoreColor}-600">${data.match_score}%</div>
+                    <div>
+                        <p class="text-sm font-medium text-slate-700">Semantic Match Score</p>
+                        <div class="w-32 h-2 bg-slate-100 rounded-full overflow-hidden mt-1">
+                            <div class="h-full rounded-full bg-${scoreColor}-500" style="width: ${data.match_score}%"></div>
+                        </div>
+                    </div>
+                </div>
+
+                <!-- Skills -->
+                <div class="mb-5">
+                    <h4 class="text-xs font-bold text-slate-500 uppercase mb-2">All Skills (${data.skills.length})</h4>
+                    <div class="flex flex-wrap gap-1.5">${skillsHtml}</div>
+                </div>
+
+                <!-- Education & Experience -->
+                <div class="grid grid-cols-1 md:grid-cols-2 gap-4 mb-5">
+                    <div>
+                        <h4 class="text-xs font-bold text-violet-600 uppercase mb-2">Education</h4>
+                        <div class="flex flex-wrap gap-1.5">${eduHtml}</div>
+                    </div>
+                    <div>
+                        <h4 class="text-xs font-bold text-teal-600 uppercase mb-2">Experience</h4>
+                        <div class="flex flex-wrap gap-1.5">${expHtml}</div>
+                    </div>
+                </div>
+
+                <!-- Skill Gap -->
+                <div class="grid grid-cols-1 md:grid-cols-2 gap-4">
+                    <div>
+                        <h4 class="text-xs font-bold text-emerald-600 uppercase mb-2">✓ Matched Skills</h4>
+                        <div class="flex flex-wrap gap-1.5">${matchedHtml}</div>
+                    </div>
+                    <div>
+                        <h4 class="text-xs font-bold text-red-600 uppercase mb-2">✗ Missing Skills</h4>
+                        <div class="flex flex-wrap gap-1.5">${missingHtml}</div>
+                    </div>
+                </div>
+
+                ${audioHtml}
+            `;
+        })
+        .catch(err => {
+            body.innerHTML = `<p class="text-red-500 text-center py-12">Error loading candidate data</p>`;
+        });
+}
+
+function closeModal() {
+    document.getElementById('candidate-modal').classList.add('hidden');
+    document.body.style.overflow = '';
+}
+
+// Close modal on ESC key
+document.addEventListener('keydown', function(e) {
+    if (e.key === 'Escape') closeModal();
+});
+</script>
+{% endblock %}

templates/register.html

@@ -0,0 +1,55 @@
+{% extends "base.html" %}
+{% block content %}
+<div class="flex-grow flex items-center justify-center px-4 py-10">
+    <div class="w-full max-w-md animate-fade-in-up">
+        <!-- Brand Header -->
+        <div class="text-center mb-8">
+            <div class="inline-flex items-center justify-center w-16 h-16 bg-gradient-to-br from-brand-500 to-sky-500 rounded-2xl shadow-lg shadow-brand-200 mb-4">
+                <svg class="w-8 h-8 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M18 9v3m0 0v3m0-3h3m-3 0h-3m-2-5a4 4 0 11-8 0 4 4 0 018 0zM3 20a6 6 0 0112 0v1H3v-1z"></path>
+                </svg>
+            </div>
+            <h1 class="text-2xl font-extrabold text-slate-800 mb-1">Create your account</h1>
+            <p class="text-slate-500 text-sm">Join HireScope AI — the next-gen screening platform</p>
+        </div>
+
+        <!-- Register Card -->
+        <div class="glass-card p-8 rounded-2xl">
+            <form method="POST" action="/register" class="space-y-5">
+                <div>
+                    <label for="username" class="block text-sm font-semibold text-slate-700 mb-1.5">Full Name</label>
+                    <input type="text" name="username" id="username" required
+                        class="w-full bg-slate-50 border border-slate-200 text-slate-800 rounded-xl px-4 py-3 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition placeholder-slate-400"
+                        placeholder="Alex Johnson">
+                </div>
+                
+                <div>
+                    <label for="email" class="block text-sm font-semibold text-slate-700 mb-1.5">Email Address</label>
+                    <input type="email" name="email" id="email" required
+                        class="w-full bg-slate-50 border border-slate-200 text-slate-800 rounded-xl px-4 py-3 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition placeholder-slate-400"
+                        placeholder="alex@company.com">
+                </div>
+                
+                <div>
+                    <label for="password" class="block text-sm font-semibold text-slate-700 mb-1.5">Password</label>
+                    <input type="password" name="password" id="password" required minlength="6"
+                        class="w-full bg-slate-50 border border-slate-200 text-slate-800 rounded-xl px-4 py-3 focus:outline-none focus:border-brand-400 focus:ring-2 focus:ring-brand-100 transition placeholder-slate-400"
+                        placeholder="Min. 6 characters">
+                </div>
+                
+                <button type="submit" class="w-full btn-primary text-white font-semibold py-3 rounded-xl shadow-lg text-sm flex items-center justify-center gap-2">
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M18 9v3m0 0v3m0-3h3m-3 0h-3m-2-5a4 4 0 11-8 0 4 4 0 018 0zM3 20a6 6 0 0112 0v1H3v-1z"></path></svg>
+                    Create Account
+                </button>
+            </form>
+        </div>
+
+        <!-- Footer Link -->
+        <div class="mt-6 text-center">
+            <p class="text-slate-500 text-sm">Already have an account? 
+                <a href="/login" class="text-brand-600 hover:text-brand-700 font-semibold transition">Sign In</a>
+            </p>
+        </div>
+    </div>
+</div>
+{% endblock %}

templates/results.html

@@ -0,0 +1,316 @@
+{% extends "base.html" %}
+{% block content %}
+<div class="max-w-6xl mx-auto w-full px-4 sm:px-6 lg:px-8 mt-2 animate-fade-in-up">
+    
+    <!-- Header -->
+    <div class="flex flex-col sm:flex-row justify-between items-start sm:items-center mb-8 gap-4">
+        <div>
+            <h1 class="text-3xl font-extrabold text-slate-800 flex items-center gap-3">
+                <div class="w-10 h-10 bg-gradient-to-br from-emerald-500 to-teal-500 rounded-xl flex items-center justify-center shadow-md shadow-emerald-200">
+                    <svg class="w-5 h-5 text-white" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12l2 2 4-4m6 2a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+                </div>
+                Analysis Results
+            </h1>
+            <p class="text-slate-500 text-sm mt-1 ml-13">AI-powered resume analysis and skill extraction</p>
+        </div>
+        <div class="flex gap-3">
+            <a href="/" class="px-4 py-2.5 bg-white hover:bg-slate-50 text-slate-700 rounded-xl transition border border-slate-200 shadow-sm font-medium text-sm flex items-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 4v16m8-8H4"></path></svg>
+                Upload Another
+            </a>
+            <a href="/ranking" class="btn-primary text-white px-4 py-2.5 rounded-xl shadow-md font-medium text-sm flex items-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 7h8m0 0v8m0-8l-8 8-4-4-6 6"></path></svg>
+                View Leaderboard
+            </a>
+        </div>
+    </div>
+
+    {% if not candidate %}
+        <!-- Empty State -->
+        <div class="glass-card p-16 text-center rounded-2xl">
+            <div class="w-20 h-20 bg-slate-100 rounded-2xl flex items-center justify-center mx-auto mb-6">
+                <svg class="w-10 h-10 text-slate-300" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M20 13V6a2 2 0 00-2-2H6a2 2 0 00-2 2v7m16 0v5a2 2 0 01-2 2H6a2 2 0 01-2-2v-5m16 0h-2.586a1 1 0 00-.707.293l-2.414 2.414a1 1 0 01-.707.293h-3.172a1 1 0 01-.707-.293l-2.414-2.414A1 1 0 006.586 13H4"></path></svg>
+            </div>
+            <h3 class="text-xl font-bold text-slate-700 mb-2">No candidate analyzed yet</h3>
+            <p class="text-slate-400 mb-6">Upload a resume to see AI-powered analysis results</p>
+            <a href="/" class="btn-primary text-white px-6 py-3 rounded-xl shadow-md font-semibold text-sm inline-flex items-center gap-2">
+                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16v1a3 3 0 003 3h10a3 3 0 003-3v-1m-4-8l-4-4m0 0L8 8m4-4v12"></path></svg>
+                Upload Resume
+            </a>
+        </div>
+    {% else %}
+    
+    <!-- Score + Candidate Info Row -->
+    <div class="grid grid-cols-1 lg:grid-cols-3 gap-6 mb-6">
+        
+        <!-- Semantic Score Card -->
+        <div class="glass-card p-6 rounded-2xl flex flex-col items-center justify-center relative overflow-hidden group animate-fade-in-up">
+            <div class="absolute inset-0 bg-gradient-to-br from-brand-50/50 to-sky-50/50 opacity-0 group-hover:opacity-100 transition duration-500"></div>
+            <h3 class="text-slate-500 font-semibold mb-2 uppercase tracking-wider text-xs relative z-10">Semantic Match</h3>
+            
+            <!-- Circular Score -->
+            <div class="relative w-36 h-36 flex items-center justify-center mt-2 z-10">
+                <svg class="w-full h-full transform -rotate-90" viewBox="0 0 36 36">
+                    <path class="text-slate-100" d="M18 2.0845 a 15.9155 15.9155 0 0 1 0 31.831 a 15.9155 15.9155 0 0 1 0 -31.831" fill="none" stroke="currentColor" stroke-width="2.5" />
+                    <path 
+                        class="{% if candidate.match_score >= 70 %}text-emerald-500{% elif candidate.match_score >= 40 %}text-amber-500{% else %}text-brand-500{% endif %}"
+                        stroke-dasharray="{{ candidate.match_score }}, 100" 
+                        d="M18 2.0845 a 15.9155 15.9155 0 0 1 0 31.831 a 15.9155 15.9155 0 0 1 0 -31.831" 
+                        fill="none" stroke="currentColor" stroke-width="2.5" 
+                        stroke-linecap="round"
+                        style="filter: drop-shadow(0 0 6px {% if candidate.match_score >= 70 %}rgba(16,185,129,0.4){% elif candidate.match_score >= 40 %}rgba(245,158,11,0.4){% else %}rgba(99,102,241,0.4){% endif %});"
+                    />
+                </svg>
+                <div class="absolute flex flex-col items-center">
+                    <span class="text-4xl font-extrabold text-slate-800">{{ candidate.match_score }}<span class="text-lg text-slate-400 font-medium">%</span></span>
+                </div>
+            </div>
+            <p class="text-center text-xs text-slate-400 mt-4 px-4 relative z-10 leading-relaxed">
+                {% if candidate.job_description %}
+                    AI Semantic Score against provided Job Description
+                {% else %}
+                    Provide a Job Description to unlock semantic scoring
+                {% endif %}
+            </p>
+        </div>
+
+        <!-- Candidate Info Card -->
+        <div class="glass-card p-6 rounded-2xl lg:col-span-2 flex flex-col animate-fade-in-up-delay">
+            <h3 class="text-slate-500 font-semibold mb-3 uppercase tracking-wider text-xs">Candidate Profile</h3>
+            
+            <div class="flex items-center gap-4 mb-5">
+                <div class="w-14 h-14 bg-gradient-to-br from-brand-500 to-sky-500 rounded-2xl flex items-center justify-center text-white text-xl font-bold shadow-lg shadow-brand-200">
+                    {{ candidate.name[0]|upper }}
+                </div>
+                <div>
+                    <h2 class="text-xl font-bold text-slate-800">{{ candidate.name }}</h2>
+                    <p class="text-xs text-slate-400">{{ candidate.skills|length }} skills • {{ candidate.education|length }} education keywords • {{ candidate.experience|length }} experience indicators</p>
+                </div>
+            </div>
+
+            {% if candidate.ai_summary %}
+            <!-- Google Gen AI Summary -->
+            <div class="bg-gradient-to-r from-blue-50 to-indigo-50 p-4 rounded-xl border border-blue-100 mb-5">
+                <div class="text-xs text-blue-600 uppercase font-bold mb-2 flex items-center gap-1.5">
+                    <svg class="w-4 h-4 text-blue-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M13 10V3L4 14h7v7l9-11h-7z"></path></svg>
+                    AI Summary Overview
+                </div>
+                <p class="text-slate-700 text-sm italic leading-relaxed">{{ candidate.ai_summary }}</p>
+            </div>
+            {% endif %}
+            
+            <div class="grid grid-cols-1 md:grid-cols-2 gap-4 flex-grow">
+                <!-- Education -->
+                <div class="bg-gradient-to-br from-violet-50 to-purple-50 p-4 rounded-xl border border-violet-100">
+                    <div class="text-xs text-violet-600 uppercase font-bold mb-2 flex items-center gap-1.5">
+                        <svg class="w-3.5 h-3.5" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 14l9-5-9-5-9 5 9 5z"></path><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 14l6.16-3.422a12.083 12.083 0 01.665 6.479A11.952 11.952 0 0012 20.055a11.952 11.952 0 00-6.824-2.998 12.078 12.078 0 01.665-6.479L12 14z"></path></svg>
+                        Education
+                    </div>
+                    {% if candidate.education %}
+                        <div class="flex flex-wrap gap-1.5">
+                            {% for e in candidate.education %}
+                            <span class="px-2.5 py-1 bg-white/80 text-violet-700 border border-violet-200 rounded-lg text-xs font-medium">{{ e }}</span>
+                            {% endfor %}
+                        </div>
+                    {% else %}
+                        <span class="text-violet-400 text-xs italic">None explicitly detected</span>
+                    {% endif %}
+                </div>
+                
+                <!-- Experience -->
+                <div class="bg-gradient-to-br from-teal-50 to-emerald-50 p-4 rounded-xl border border-teal-100">
+                    <div class="text-xs text-teal-600 uppercase font-bold mb-2 flex items-center gap-1.5">
+                        <svg class="w-3.5 h-3.5" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M21 13.255A23.931 23.931 0 0112 15c-3.183 0-6.22-.62-9-1.745M16 6V4a2 2 0 00-2-2h-4a2 2 0 00-2 2v2m4 6h.01M5 20h14a2 2 0 002-2V8a2 2 0 00-2-2H5a2 2 0 00-2 2v10a2 2 0 002 2z"></path></svg>
+                        Experience
+                    </div>
+                    {% if candidate.experience %}
+                        <div class="flex flex-wrap gap-1.5">
+                            {% for ex in candidate.experience %}
+                            <span class="px-2.5 py-1 bg-white/80 text-teal-700 border border-teal-200 rounded-lg text-xs font-medium">{{ ex }}</span>
+                            {% endfor %}
+                        </div>
+                    {% else %}
+                        <span class="text-teal-400 text-xs italic">None explicitly detected</span>
+                    {% endif %}
+                </div>
+            </div>
+        </div>
+    </div>
+    
+    <!-- All Skills Section -->
+    <div class="glass-card p-6 rounded-2xl mb-6 animate-fade-in-up-delay">
+        <h3 class="text-slate-500 font-semibold mb-4 uppercase tracking-wider text-xs flex items-center gap-2">
+            <svg class="w-4 h-4 text-brand-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M7 7h.01M7 3h5c.512 0 1.024.195 1.414.586l7 7a2 2 0 010 2.828l-7 7a2 2 0 01-2.828 0l-7-7A1.994 1.994 0 013 12V7a4 4 0 014-4z"></path></svg>
+            All Extracted Skills ({{ candidate.skills|length }})
+        </h3>
+        <div class="flex flex-wrap gap-2">
+            {% for skill in candidate.skills %}
+                <span class="skill-tag px-3 py-1.5 bg-gradient-to-r from-brand-50 to-sky-50 text-brand-700 border border-brand-200 rounded-full text-xs font-semibold cursor-default">{{ skill }}</span>
+            {% endfor %}
+            {% if not candidate.skills %}
+                <span class="text-slate-400 text-sm italic">No skills detected</span>
+            {% endif %}
+        </div>
+    </div>
+
+    <!-- Skills Gap Analysis -->
+    <div class="grid grid-cols-1 lg:grid-cols-2 gap-6 mb-6">
+        <!-- Matched Skills -->
+        <div class="glass-card p-6 rounded-2xl animate-fade-in-up-delay">
+            <h3 class="text-slate-500 font-semibold mb-4 uppercase tracking-wider text-xs flex items-center gap-2">
+                <span class="w-2.5 h-2.5 rounded-full bg-emerald-400 shadow-sm shadow-emerald-200"></span>
+                Matched Skills (JD)
+            </h3>
+            <div class="flex flex-wrap gap-2">
+                {% if candidate.skill_gaps.matched %}
+                    {% for m in candidate.skill_gaps.matched %}
+                    <span class="skill-tag px-3 py-1.5 bg-emerald-50 text-emerald-700 border border-emerald-200 rounded-full font-semibold text-xs">✓ {{ m }}</span>
+                    {% endfor %}
+                {% elif not candidate.job_description %}
+                    <span class="text-slate-400 text-sm italic">Provide a Job Description to see skill matches</span>
+                {% else %}
+                    <span class="text-slate-400 text-sm italic">No exact skill overlaps found</span>
+                {% endif %}
+            </div>
+        </div>
+        
+        <!-- Missing Skills -->
+        <div class="glass-card p-6 rounded-2xl animate-fade-in-up-delay-2">
+             <h3 class="text-slate-500 font-semibold mb-4 uppercase tracking-wider text-xs flex items-center gap-2">
+                <span class="w-2.5 h-2.5 rounded-full bg-red-400 shadow-sm shadow-red-200"></span>
+                Missing Skills (JD)
+            </h3>
+            <div class="flex flex-wrap gap-2">
+                {% if candidate.skill_gaps.missing %}
+                    {% for m in candidate.skill_gaps.missing %}
+                    <span class="skill-tag px-3 py-1.5 bg-red-50 text-red-600 border border-red-200 rounded-full font-semibold text-xs line-through decoration-red-300">{{ m }}</span>
+                    {% endfor %}
+                {% elif not candidate.job_description %}
+                    <span class="text-slate-400 text-sm italic">No Job Description provided for gap analysis</span>
+                {% else %}
+                    <span class="text-emerald-600 text-sm font-medium">🎉 No skill gaps! All JD keywords present.</span>
+                {% endif %}
+            </div>
+        </div>
+    </div>
+
+    <!-- Audio Transcription Section -->
+    <div id="audio-section" class="glass-card p-6 rounded-2xl mb-6 animate-fade-in-up-delay-2 relative overflow-hidden
+        {% if transcription_pending %}border-l-4 border-amber-400
+        {% elif candidate.audio_transcription and candidate.audio_transcription.error %}border-l-4 border-red-400
+        {% elif candidate.audio_transcription and candidate.audio_transcription.text %}border-l-4 border-emerald-400
+        {% else %}border-l-4 border-brand-400{% endif %}">
+        
+        {% if transcription_pending %}
+            <!-- Processing State -->
+            <div class="flex items-start gap-4">
+                <div class="w-10 h-10 bg-amber-100 rounded-xl flex items-center justify-center flex-shrink-0">
+                    <div class="w-5 h-5 border-2 border-amber-500 border-t-transparent rounded-full animate-spin"></div>
+                </div>
+                <div class="flex-grow">
+                    <h3 class="text-amber-700 font-bold text-sm mb-1">Transcription in Progress</h3>
+                    <p class="text-slate-500 text-sm">Your audio is being transcribed with Whisper AI. This usually takes 30-60 seconds.</p>
+                    <div class="mt-3 bg-amber-50 rounded-lg p-3 border border-amber-100">
+                        <div class="flex items-center gap-2">
+                            <div class="flex gap-1">
+                                <div class="w-2 h-2 bg-amber-400 rounded-full animate-bounce" style="animation-delay: 0s"></div>
+                                <div class="w-2 h-2 bg-amber-400 rounded-full animate-bounce" style="animation-delay: 0.1s"></div>
+                                <div class="w-2 h-2 bg-amber-400 rounded-full animate-bounce" style="animation-delay: 0.2s"></div>
+                            </div>
+                            <span class="text-amber-600 text-xs font-medium" id="poll-status">Checking transcription status...</span>
+                        </div>
+                    </div>
+                </div>
+            </div>
+            <script>
+                // AJAX polling instead of full page reload
+                const candidateId = "{{ candidate._id }}";
+                function pollTranscription() {
+                    fetch(`/api/transcription_status/${candidateId}`)
+                        .then(r => r.json())
+                        .then(data => {
+                            if (data.status === 'completed' || data.status === 'failed') {
+                                window.location.reload();
+                            } else {
+                                document.getElementById('poll-status').textContent = 'Still processing... checking again in 4s';
+                                setTimeout(pollTranscription, 4000);
+                            }
+                        })
+                        .catch(() => setTimeout(pollTranscription, 5000));
+                }
+                setTimeout(pollTranscription, 3000);
+            </script>
+
+        {% elif candidate.audio_transcription and candidate.audio_transcription.error %}
+            <!-- Error State -->
+            <div class="flex items-start gap-4">
+                <div class="w-10 h-10 bg-red-100 rounded-xl flex items-center justify-center flex-shrink-0">
+                    <svg class="w-5 h-5 text-red-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 8v4m0 4h.01M21 12a9 9 0 11-18 0 9 9 0 0118 0z"></path></svg>
+                </div>
+                <div class="flex-grow">
+                    <h3 class="text-red-700 font-bold text-sm mb-1">Transcription Failed</h3>
+                    <p class="text-red-600 text-sm bg-red-50 p-3 rounded-lg border border-red-100 mt-2">{{ candidate.audio_transcription.error }}</p>
+                    
+                    <!-- Retry Upload -->
+                    <div class="mt-4">
+                        <p class="text-slate-500 text-xs mb-2">Try uploading the audio again:</p>
+                        <form action="/upload_audio" method="POST" enctype="multipart/form-data" class="flex flex-col sm:flex-row gap-3">
+                            <input type="hidden" name="candidate_id" value="{{ candidate._id }}">
+                            <div class="flex-grow">
+                                <input type="file" name="audio" required accept="audio/*" class="w-full text-sm text-slate-500 file:mr-4 file:py-2 file:px-4 file:rounded-lg file:border-0 file:text-sm file:font-semibold file:bg-brand-50 file:text-brand-600 hover:file:bg-brand-100 transition cursor-pointer">
+                            </div>
+                            <button type="submit" class="btn-secondary text-white px-4 py-2 rounded-xl shadow-sm font-medium text-sm whitespace-nowrap flex items-center gap-2">
+                                <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 4v5h.582m15.356 2A8.001 8.001 0 004.582 9m0 0H9m11 11v-5h-.581m0 0a8.003 8.003 0 01-15.357-2m15.357 2H15"></path></svg>
+                                Retry Transcription
+                            </button>
+                        </form>
+                    </div>
+                </div>
+            </div>
+
+        {% elif candidate.audio_transcription and candidate.audio_transcription.text %}
+            <!-- Success State -->
+            <div class="flex items-start gap-4">
+                <div class="w-10 h-10 bg-emerald-100 rounded-xl flex items-center justify-center flex-shrink-0">
+                    <svg class="w-5 h-5 text-emerald-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 11a7 7 0 01-7 7m0 0a7 7 0 01-7-7m7 7v4m0 0H8m4 0h4m-4-8a3 3 0 01-3-3V5a3 3 0 116 0v6a3 3 0 01-3 3z"></path></svg>
+                </div>
+                <div class="flex-grow">
+                    <h3 class="text-emerald-700 font-bold text-sm mb-1 flex items-center gap-2">
+                        Audio Transcription Complete
+                        <span class="px-2 py-0.5 bg-emerald-50 text-emerald-600 rounded-full text-xs font-medium border border-emerald-200">{{ candidate.audio_transcription.language|upper }}</span>
+                    </h3>
+                    <div class="mt-3 bg-gradient-to-r from-slate-50 to-emerald-50/50 p-4 rounded-xl border border-slate-200">
+                        <p class="text-slate-700 leading-relaxed text-sm italic">"{{ candidate.audio_transcription.text }}"</p>
+                    </div>
+                </div>
+            </div>
+
+        {% else %}
+            <!-- Upload Audio State -->
+            <div class="flex items-start gap-4">
+                <div class="w-10 h-10 bg-brand-100 rounded-xl flex items-center justify-center flex-shrink-0">
+                    <svg class="w-5 h-5 text-brand-500" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 11a7 7 0 01-7 7m0 0a7 7 0 01-7-7m7 7v4m0 0H8m4 0h4m-4-8a3 3 0 01-3-3V5a3 3 0 116 0v6a3 3 0 01-3 3z"></path></svg>
+                </div>
+                <div class="flex-grow">
+                    <h3 class="text-brand-700 font-bold text-sm mb-1">Attach Audio Introduction</h3>
+                    <p class="text-slate-500 text-sm mb-4">Upload a short audio intro from this candidate. Whisper AI will transcribe it automatically.</p>
+                    
+                    <form action="/upload_audio" method="POST" enctype="multipart/form-data" class="flex flex-col sm:flex-row gap-3">
+                        <input type="hidden" name="candidate_id" value="{{ candidate._id }}">
+                        <div class="flex-grow">
+                            <input id="audio-file-input" type="file" name="audio" required accept="audio/*" class="w-full text-sm text-slate-500 file:mr-4 file:py-2 file:px-4 file:rounded-lg file:border-0 file:text-sm file:font-semibold file:bg-brand-50 file:text-brand-600 hover:file:bg-brand-100 transition cursor-pointer">
+                        </div>
+                        <button type="submit" class="btn-secondary text-white px-5 py-2.5 rounded-xl shadow-md font-semibold text-sm whitespace-nowrap flex items-center gap-2">
+                            <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 11a7 7 0 01-7 7m0 0a7 7 0 01-7-7m7 7v4m0 0H8m4 0h4m-4-8a3 3 0 01-3-3V5a3 3 0 116 0v6a3 3 0 01-3 3z"></path></svg>
+                            Transcribe Audio
+                        </button>
+                    </form>
+                    <p class="text-xs text-slate-400 mt-2">Supported: MP3, WAV, M4A, FLAC, OGG, WEBM</p>
+                </div>
+            </div>
+        {% endif %}
+    </div>
+
+    {% endif %}
+</div>
+{% endblock %}