Deploy ProfillyBot: Gradio ZeroGPU + CV RAG (Qwen2.5-3B)

.dockerignore

@@ -0,0 +1,48 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+dist/
+*.egg-info/
+.venv/
+venv/
+env/
+
+# UV
+.uv/
+uv.lock
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# Git
+.git/
+.gitignore
+
+# Logs
+*.log
+logs/
+
+# Vector database and indexes (will be built in container)
+chroma_db/
+bm25_index/
+
+# Testing
+.pytest_cache/
+.coverage
+htmlcov/
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Documentation
+*.md
+!README_HF.md
+docs/

.env.template

@@ -0,0 +1,55 @@
+# Environment Configuration
+# Copy this file to .env and fill in your values
+
+# Application Settings
+APP_NAME="Profile Chatbot"
+APP_PORT=7860
+DEBUG=false
+
+# =============================================================================
+# Ollama Configuration (Local LLM)
+# =============================================================================
+OLLAMA_BASE_URL=http://localhost:11434
+OLLAMA_HOST=http://localhost:11434
+
+# Model to use with Ollama (pulled automatically on first run with Docker Compose)
+# Options: llama3.2:3b (default), phi3:mini, gemma2:2b, llama3.2:1b (smaller)
+OLLAMA_MODEL=llama3.2:3b
+
+# =============================================================================
+# Groq API Configuration (Cloud LLM - Optional)
+# =============================================================================
+# If set, Groq will be used as the default LLM provider instead of Ollama
+# Get your API key at: https://console.groq.com/
+GROQ_API_KEY=
+
+# Optional: Other API Keys for alternative LLM providers
+# OPENAI_API_KEY=your_openai_api_key_here
+# ANTHROPIC_API_KEY=your_anthropic_api_key_here
+
+# =============================================================================
+# Storage Configuration
+# =============================================================================
+# Vector Database
+CHROMA_PERSIST_DIR=./chroma_db
+
+# Document Processing
+DOCUMENTS_DIR=./data/documents
+
+# =============================================================================
+# Logging & Debugging
+# =============================================================================
+LOG_LEVEL=INFO
+
+# LangSmith Tracing (get API key at https://smith.langchain.com)
+# Set LANGCHAIN_TRACING_V2=true to enable tracing
+LANGCHAIN_TRACING_V2=false
+LANGCHAIN_ENDPOINT=https://api.smith.langchain.com
+LANGCHAIN_API_KEY=
+LANGCHAIN_PROJECT=profillybot
+
+# =============================================================================
+# Hugging Face Spaces (Optional - auto-set by HF Spaces)
+# =============================================================================
+# SPACE_ID=your_space_id
+# SPACE_AUTHOR_NAME=your_username

.gitattributes

@@ -33,3 +33,11 @@ saved_model/**/* 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
+.venv/Scripts/python.exe filter=lfs diff=lfs merge=lfs -text
+.venv/Scripts/pythonw.exe filter=lfs diff=lfs merge=lfs -text
+chroma_db/chroma.sqlite3 filter=lfs diff=lfs merge=lfs -text
+data/documents/MY_CV_2_0.pdf filter=lfs diff=lfs merge=lfs -text
+docs/arch_system_design.png filter=lfs diff=lfs merge=lfs -text
+docs/banner_image.png filter=lfs diff=lfs merge=lfs -text
+docs/MY_CV_2_0.pdf filter=lfs diff=lfs merge=lfs -text
+docs/screenshot-example.png filter=lfs diff=lfs merge=lfs -text

.github/workflows/README.md

@@ -0,0 +1,207 @@
+# GitHub Actions Workflows
+
+This directory contains automated CI/CD workflows for the ProfillyBot project.
+
+![Python 3.10](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python310.yml/badge.svg)
+![Python 3.11](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python311.yml/badge.svg)
+![Python 3.12](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python312.yml/badge.svg)
+![Docker](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-docker.yml/badge.svg)
+![Lint](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/lint.yml/badge.svg)
+![Code Quality](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/code-quality.yml/badge.svg)
+![Security Scan](https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/security-scan.yml/badge.svg)
+[![codecov](https://codecov.io/gh/Tin-Hoang/slm-profile-rag/graph/badge.svg?token=MW4RD9R66J)](https://codecov.io/gh/Tin-Hoang/slm-profile-rag)
+
+## Workflows Overview
+
+### 🧪 test-python3XX.yml - Unit Tests per Python Version
+**Triggers:** Push/PR to main, master, develop branches; Manual dispatch
+
+**Purpose:** Runs comprehensive unit tests for specific Python version (separate badges!)
+
+**Workflows:**
+- **test-python310.yml**: Tests on Python 3.10
+- **test-python311.yml**: Tests on Python 3.11 (includes Codecov upload)
+- **test-python312.yml**: Tests on Python 3.12
+
+**Implementation:**
+- Each workflow calls `test-python-reusable.yml` (DRY principle)
+- Installs dependencies
+- Runs ruff linting and formatting checks
+- Executes pytest with coverage reporting
+- Provides separate badge per Python version
+
+### 🐳 test-docker.yml - Docker Build Test
+**Triggers:** Push/PR to main, master, develop branches; Manual dispatch
+
+**Purpose:** Validates Docker image build
+
+**Jobs:**
+- Builds Docker image to ensure Dockerfile is valid
+- Uses caching for faster builds
+
+### 🔍 lint.yml - Linting (Existing)
+**Triggers:** Push/PR to main, develop branches
+
+**Purpose:** Fast linting checks using ruff
+
+**Jobs:**
+- Runs ruff check on all Python files
+- Runs ruff format check
+- Uses UV for fast dependency installation
+
+### 📊 code-quality.yml - Code Quality Analysis
+**Triggers:** Push/PR to main, master, develop branches
+
+**Purpose:** Advanced code quality checks
+
+**Jobs:**
+- Python syntax validation
+- Code complexity analysis using radon
+- Maintainability index calculation
+
+### ✅ pre-commit.yml - Pre-commit Hooks
+**Triggers:** Push/PR to main, master, develop branches
+
+**Purpose:** Validates pre-commit hooks configuration
+
+**Jobs:**
+- Runs all pre-commit hooks defined in `.pre-commit-config.yaml`
+- Ensures code formatting and quality standards
+
+### 📦 dependency-review.yml - Dependency Security
+**Triggers:** Pull requests only
+
+**Purpose:** Reviews dependency changes for security issues
+
+**Jobs:**
+- Scans for vulnerable dependencies
+- Checks for problematic licenses (GPL-3.0, AGPL-3.0)
+- Fails on moderate or higher severity vulnerabilities
+
+### 🔒 security-scan.yml - Security Scanning
+**Triggers:** Push/PR to main, master, develop branches; Weekly schedule (Mondays)
+
+**Purpose:** Comprehensive security vulnerability scanning
+
+**Jobs:**
+- **Safety**: Scans Python dependencies for known vulnerabilities
+- **Bandit**: Static security analysis of Python code
+- **pip-audit**: Audits Python packages for known vulnerabilities
+- Uploads security reports as artifacts
+
+## Local Development
+
+### Running Tests Locally
+
+```bash
+# Install development dependencies
+pip install -e ".[dev]"
+
+# Run all tests
+pytest tests/ -v
+
+# Run tests with coverage
+pytest tests/ -v --cov=src --cov-report=term-missing --cov-report=html
+
+# Run specific test file
+pytest tests/test_main_document_loader.py -v
+```
+
+### Running Linting Locally
+
+```bash
+# Install ruff
+pip install ruff
+
+# Check code
+ruff check src tests app.py
+
+# Format code
+ruff format src tests app.py
+```
+
+### Running Pre-commit Hooks Locally
+
+```bash
+# Install pre-commit
+pip install pre-commit
+
+# Install hooks
+pre-commit install
+
+# Run on all files
+pre-commit run --all-files
+```
+
+### Running Security Scans Locally
+
+```bash
+# Install security tools
+pip install safety bandit pip-audit
+
+# Run safety check
+safety check
+
+# Run bandit
+bandit -r src/
+
+# Run pip-audit
+pip-audit
+```
+
+## Continuous Integration Matrix
+
+| Workflow | Python Versions | OS | Caching | Coverage |
+|----------|----------------|-----|---------|----------|
+| test-python310.yml | 3.10 | Ubuntu | ✅ pip | ❌ |
+| test-python311.yml | 3.11 | Ubuntu | ✅ pip | ✅ Codecov |
+| test-python312.yml | 3.12 | Ubuntu | ✅ pip | ❌ |
+| test-docker.yml | N/A | Ubuntu | ✅ Docker | ❌ |
+| lint.yml | 3.11 | Ubuntu | ✅ UV | ❌ |
+| code-quality.yml | 3.11 | Ubuntu | ✅ pip | ❌ |
+| pre-commit.yml | 3.11 | Ubuntu | ✅ pre-commit | ❌ |
+| security-scan.yml | 3.11 | Ubuntu | ✅ pip | ❌ |
+
+## Troubleshooting
+
+### Tests Failing Locally but Passing in CI (or vice versa)
+
+1. Ensure you're using the correct Python version
+2. Clear pytest cache: `pytest --cache-clear`
+3. Reinstall dependencies: `pip install -e ".[dev]" --force-reinstall`
+
+### Codecov Upload Issues
+
+1. Ensure `CODECOV_TOKEN` is set in repository secrets
+2. Check coverage.xml is generated: `pytest tests/ --cov=src --cov-report=xml`
+
+### Docker Build Failures
+
+1. Test locally: `docker build -t profillybot:test .`
+2. Check Dockerfile syntax
+3. Ensure all required files are present
+
+## Contributing
+
+When contributing to this project:
+
+1. Ensure all workflows pass before submitting a PR
+2. Add tests for new features
+3. Run pre-commit hooks locally
+4. Update this README if adding new workflows
+
+## Maintenance
+
+### Updating Dependencies
+
+- GitHub Actions are set to use latest stable versions (v4, v5)
+- Review and update action versions quarterly
+- Monitor Dependabot alerts for security updates
+
+### Adding New Workflows
+
+1. Create a new `.yml` file in this directory
+2. Follow the naming convention: `<purpose>.yml`
+3. Add appropriate triggers and jobs
+4. Update this README with workflow documentation
+5. Add status badge to main README.md

.github/workflows/code-quality.yml

@@ -0,0 +1,38 @@
+name: Code Quality
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+
+jobs:
+  type-check:
+    name: Type Checking & Complexity Analysis
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.11"
+        cache: 'pip'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install mypy radon
+
+    - name: Check for Python syntax errors
+      run: |
+        python -m py_compile app.py
+        find src tests -name "*.py" -exec python -m py_compile {} \;
+
+    - name: Run complexity analysis
+      continue-on-error: true
+      run: |
+        radon cc src tests app.py -a -s
+        radon mi src tests app.py -s

.github/workflows/dependency-review.yml

@@ -0,0 +1,22 @@
+name: Dependency Review
+
+on:
+  pull_request:
+    branches: [ main, master, develop ]
+
+permissions:
+  contents: read
+
+jobs:
+  dependency-review:
+    name: Review Dependencies
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Dependency Review
+        uses: actions/dependency-review-action@v4
+        with:
+          fail-on-severity: moderate
+          deny-licenses: GPL-3.0, AGPL-3.0

.github/workflows/lint.yml

@@ -0,0 +1,38 @@
+name: Lint
+
+on:
+  push:
+    branches: [ main, develop ]
+  pull_request:
+    branches: [ main, develop ]
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.11'
+
+    - name: Install UV
+      run: curl -LsSf https://astral.sh/uv/install.sh | sh
+
+    - name: Install dependencies
+      run: |
+        uv venv
+        source .venv/bin/activate
+        uv pip install -r requirements.txt
+        uv pip install "ruff>=0.14.6,<0.15.0" pytest pytest-cov
+
+    - name: Run Ruff Check
+      run: |
+        source .venv/bin/activate
+        ruff check .
+
+    - name: Run Ruff Format Check
+      run: |
+        source .venv/bin/activate
+        ruff format --check .

.github/workflows/pre-commit.yml

@@ -0,0 +1,38 @@
+name: Pre-commit Checks
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+
+jobs:
+  pre-commit:
+    name: Run Pre-commit Hooks
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.10"
+
+    - name: Install pre-commit
+      run: |
+        python -m pip install --upgrade pip
+        pip install pre-commit
+
+    - name: Cache pre-commit hooks
+      uses: actions/cache@v4
+      with:
+        path: ~/.cache/pre-commit
+        key: pre-commit-${{ hashFiles('.pre-commit-config.yaml') }}
+        restore-keys: |
+          pre-commit-
+
+    - name: Run pre-commit on all files
+      run: |
+        pre-commit run --all-files --show-diff-on-failure

.github/workflows/security-scan.yml

@@ -0,0 +1,55 @@
+name: Security Scan
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+  schedule:
+    # Run weekly on Monday at 00:00 UTC
+    - cron: '0 0 * * 1'
+
+jobs:
+  security-scan:
+    name: Security Vulnerability Scan
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.11"
+        cache: 'pip'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install safety bandit pip-audit
+
+    - name: Run Safety check
+      continue-on-error: true
+      run: |
+        pip install -e .
+        safety check --json || true
+
+    - name: Run Bandit security scan
+      continue-on-error: true
+      run: |
+        bandit -r src/ -f json -o bandit-report.json || true
+        bandit -r src/ -f txt || true
+
+    - name: Run pip-audit
+      continue-on-error: true
+      run: |
+        pip-audit --desc || true
+
+    - name: Upload Bandit report
+      if: always()
+      uses: actions/upload-artifact@v4
+      with:
+        name: bandit-security-report
+        path: bandit-report.json
+        if-no-files-found: ignore

.github/workflows/test-docker.yml

@@ -0,0 +1,28 @@
+name: Docker Build Test
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+  workflow_dispatch:
+
+jobs:
+  test-docker:
+    name: Test Docker Build
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v3
+
+    - name: Build Docker image
+      uses: docker/build-push-action@v5
+      with:
+        context: .
+        push: false
+        tags: profillybot:test
+        cache-from: type=gha
+        cache-to: type=gha,mode=max

.github/workflows/test-python-reusable.yml

@@ -0,0 +1,48 @@
+name: Test Python (Reusable)
+
+on:
+  workflow_call:
+    inputs:
+      python-version:
+        required: true
+        type: string
+    secrets:
+      CODECOV_TOKEN:
+        required: false
+
+jobs:
+  test:
+    name: Test on Python ${{ inputs.python-version }}
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Python ${{ inputs.python-version }}
+      uses: actions/setup-python@v5
+      with:
+        python-version: ${{ inputs.python-version }}
+        cache: 'pip'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install -e ".[dev]"
+
+    - name: Run linting with ruff
+      run: |
+        ruff check src tests
+        ruff format --check src tests
+
+    - name: Run tests with pytest
+      run: |
+        pytest tests/ -v --cov=src --cov-report=term-missing --cov-report=xml
+
+    - name: Upload coverage to Codecov
+      if: inputs.python-version == '3.11'
+      uses: codecov/codecov-action@v4
+      with:
+        file: ./coverage.xml
+        token: ${{ secrets.CODECOV_TOKEN }}
+        fail_ci_if_error: false
+        verbose: true

.github/workflows/test-python310.yml

@@ -0,0 +1,14 @@
+name: Tests Python 3.10
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+  workflow_dispatch:
+
+jobs:
+  test:
+    uses: ./.github/workflows/test-python-reusable.yml
+    with:
+      python-version: "3.10"

.github/workflows/test-python311.yml

@@ -0,0 +1,15 @@
+name: Tests Python 3.11
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+  workflow_dispatch:
+
+jobs:
+  test:
+    uses: ./.github/workflows/test-python-reusable.yml
+    with:
+      python-version: "3.11"
+    secrets: inherit

.github/workflows/test-python312.yml

@@ -0,0 +1,14 @@
+name: Tests Python 3.12
+
+on:
+  push:
+    branches: [ main, master, develop ]
+  pull_request:
+    branches: [ main, master, develop ]
+  workflow_dispatch:
+
+jobs:
+  test:
+    uses: ./.github/workflows/test-python-reusable.yml
+    with:
+      python-version: "3.12"

.gitignore

@@ -0,0 +1,207 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+#poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+#pdm.lock
+#pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+#pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Abstra
+# Abstra is an AI-powered process automation framework.
+# Ignore directories containing user credentials, local state, and settings.
+# Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore
+#  that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#  and can be added to the global gitignore or merged into this file. However, if you prefer,
+#  you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Cursor
+#  Cursor is an AI-powered code editor. `.cursorignore` specifies files/directories to
+#  exclude from AI features like autocomplete and code analysis. Recommended for sensitive data
+#  refer to https://docs.cursor.com/context/ignore-files
+.cursorignore
+.cursorindexingignore
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/

.hfignore

@@ -0,0 +1,19 @@
+.venv/
+venv/
+.env
+.env.*
+.git/
+__pycache__/
+**/__pycache__/
+*.pyc
+.pytest_cache/
+.ruff_cache/
+htmlcov/
+.coverage
+.cursor/
+*.egg-info/
+.mypy_cache/
+_cv_preview.txt
+.idea/
+.vscode/
+agent-transcripts/

.pre-commit-config.yaml

@@ -0,0 +1,40 @@
+# See https://pre-commit.com for more information
+# See https://pre-commit.com/hooks.html for more hooks
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v4.5.0
+    hooks:
+      - id: trailing-whitespace
+      - id: end-of-file-fixer
+      - id: check-yaml
+      - id: check-toml
+      - id: check-added-large-files
+        args: ['--maxkb=6000']
+      - id: check-merge-conflict
+      - id: check-case-conflict
+      - id: check-docstring-first
+      - id: debug-statements
+      - id: mixed-line-ending
+        args: ['--fix=lf']
+
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.7.4
+    hooks:
+      # Run the linter
+      - id: ruff
+        args: ['--fix']
+      # Run the formatter
+      - id: ruff-format
+
+  # Uncomment to enable mypy type checking (requires adding mypy and type stubs to dev dependencies)
+  # - repo: https://github.com/pre-commit/mirrors-mypy
+  #   rev: v1.11.2
+  #   hooks:
+  #     - id: mypy
+  #       additional_dependencies: ['types-PyYAML']
+  #       args: ['--ignore-missing-imports', '--no-strict-optional']
+  #       exclude: '^tests/'
+
+# Set this to allow pre-commit to work with uv or other package managers
+default_language_version:
+  python: python3.10

.streamlit/config.toml

@@ -0,0 +1,16 @@
+[theme]
+primaryColor = "#FF4B4B"
+backgroundColor = "#FFFFFF"
+secondaryBackgroundColor = "#F0F2F6"
+textColor = "#262730"
+font = "sans serif"
+
+[server]
+headless = true
+# port = 8501  # Default Streamlit port (for local dev)
+# Note: HuggingFace Spaces overrides this to 7860 via Dockerfile
+enableCORS = false
+enableXsrfProtection = true
+
+[browser]
+gatherUsageStats = false

.venv/.gitignore

@@ -0,0 +1 @@
+*

.venv/CACHEDIR.TAG

@@ -0,0 +1 @@
+Signature: 8a477f597d28d172789f06886806bc55

.venv/Lib/site-packages/_virtualenv.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69ac3d8f27e679c81b94ab30b3b56e9cd138219b1ba94a1fa3606d5a76a1433d
+size 18

.venv/Lib/site-packages/_virtualenv.py

@@ -0,0 +1,101 @@
+"""Patches that are applied at runtime to the virtual environment."""
+
+import os
+import sys
+
+VIRTUALENV_PATCH_FILE = os.path.join(__file__)
+
+
+def patch_dist(dist):
+    """
+    Distutils allows user to configure some arguments via a configuration file:
+    https://docs.python.org/3.11/install/index.html#distutils-configuration-files.
+
+    Some of this arguments though don't make sense in context of the virtual environment files, let's fix them up.
+    """  # noqa: D205
+    # we cannot allow some install config as that would get packages installed outside of the virtual environment
+    old_parse_config_files = dist.Distribution.parse_config_files
+
+    def parse_config_files(self, *args, **kwargs):
+        result = old_parse_config_files(self, *args, **kwargs)
+        install = self.get_option_dict("install")
+
+        if "prefix" in install:  # the prefix governs where to install the libraries
+            install["prefix"] = VIRTUALENV_PATCH_FILE, os.path.abspath(sys.prefix)
+        for base in ("purelib", "platlib", "headers", "scripts", "data"):
+            key = f"install_{base}"
+            if key in install:  # do not allow global configs to hijack venv paths
+                install.pop(key, None)
+        return result
+
+    dist.Distribution.parse_config_files = parse_config_files
+
+
+# Import hook that patches some modules to ignore configuration values that break package installation in case
+# of virtual environments.
+_DISTUTILS_PATCH = "distutils.dist", "setuptools.dist"
+# https://docs.python.org/3/library/importlib.html#setting-up-an-importer
+
+
+class _Finder:
+    """A meta path finder that allows patching the imported distutils modules."""
+
+    fullname = None
+
+    # lock[0] is threading.Lock(), but initialized lazily to avoid importing threading very early at startup,
+    # because there are gevent-based applications that need to be first to import threading by themselves.
+    # See https://github.com/pypa/virtualenv/issues/1895 for details.
+    lock = []  # noqa: RUF012
+
+    def find_spec(self, fullname, path, target=None):  # noqa: ARG002
+        if fullname in _DISTUTILS_PATCH and self.fullname is None:
+            # initialize lock[0] lazily
+            if len(self.lock) == 0:
+                import threading
+
+                lock = threading.Lock()
+                # there is possibility that two threads T1 and T2 are simultaneously running into find_spec,
+                # observing .lock as empty, and further going into hereby initialization. However due to the GIL,
+                # list.append() operation is atomic and this way only one of the threads will "win" to put the lock
+                # - that every thread will use - into .lock[0].
+                # https://docs.python.org/3/faq/library.html#what-kinds-of-global-value-mutation-are-thread-safe
+                self.lock.append(lock)
+
+            from functools import partial
+            from importlib.util import find_spec
+
+            with self.lock[0]:
+                self.fullname = fullname
+                try:
+                    spec = find_spec(fullname, path)
+                    if spec is not None:
+                        # https://www.python.org/dev/peps/pep-0451/#how-loading-will-work
+                        is_new_api = hasattr(spec.loader, "exec_module")
+                        func_name = "exec_module" if is_new_api else "load_module"
+                        old = getattr(spec.loader, func_name)
+                        func = self.exec_module if is_new_api else self.load_module
+                        if old is not func:
+                            try:  # noqa: SIM105
+                                setattr(spec.loader, func_name, partial(func, old))
+                            except AttributeError:
+                                pass  # C-Extension loaders are r/o such as zipimporter with <3.7
+                        return spec
+                finally:
+                    self.fullname = None
+        return None
+
+    @staticmethod
+    def exec_module(old, module):
+        old(module)
+        if module.__name__ in _DISTUTILS_PATCH:
+            patch_dist(module)
+
+    @staticmethod
+    def load_module(old, name):
+        module = old(name)
+        if module.__name__ in _DISTUTILS_PATCH:
+            patch_dist(module)
+        return module
+
+
+sys.meta_path.insert(0, _Finder())

.venv/Scripts/activate

@@ -0,0 +1,116 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+# This file must be used with "source bin/activate" *from bash*
+# you cannot run it directly
+
+# Get script path (only used if environment is relocatable).
+if [ -n "${BASH_VERSION:+x}" ] ; then
+    SCRIPT_PATH="${BASH_SOURCE[0]}"
+    if [ "$SCRIPT_PATH" = "$0" ]; then
+        # Only bash has a reasonably robust check for source'dness.
+        echo "You must source this script: \$ source $0" >&2
+        exit 33
+    fi
+elif [ -n "${ZSH_VERSION:+x}" ] ; then
+    SCRIPT_PATH="${(%):-%x}"
+elif [ -n "${KSH_VERSION:+x}" ] ; then
+    SCRIPT_PATH="${.sh.file}"
+fi
+
+deactivate () {
+    unset -f pydoc >/dev/null 2>&1 || true
+
+    # reset old environment variables
+    # ! [ -z ${VAR+_} ] returns true if VAR is declared at all
+    if ! [ -z "${_OLD_VIRTUAL_PATH:+_}" ] ; then
+        PATH="$_OLD_VIRTUAL_PATH"
+        export PATH
+        unset _OLD_VIRTUAL_PATH
+    fi
+    if ! [ -z "${_OLD_VIRTUAL_PYTHONHOME+_}" ] ; then
+        PYTHONHOME="$_OLD_VIRTUAL_PYTHONHOME"
+        export PYTHONHOME
+        unset _OLD_VIRTUAL_PYTHONHOME
+    fi
+
+    # The hash command must be called to get it to forget past
+    # commands. Without forgetting past commands the $PATH changes
+    # we made may not be respected
+    hash -r 2>/dev/null
+
+    if ! [ -z "${_OLD_VIRTUAL_PS1+_}" ] ; then
+        PS1="$_OLD_VIRTUAL_PS1"
+        export PS1
+        unset _OLD_VIRTUAL_PS1
+    fi
+
+    unset VIRTUAL_ENV
+    unset VIRTUAL_ENV_PROMPT
+    if [ ! "${1-}" = "nondestructive" ] ; then
+    # Self destruct!
+        unset -f deactivate
+    fi
+}
+
+# unset irrelevant variables
+deactivate nondestructive
+
+VIRTUAL_ENV='C:\Users\Admin\Downloads\AIDevMinh\slm-profile-rag\.venv'
+if ([ "$OSTYPE" = "cygwin" ] || [ "$OSTYPE" = "msys" ]) && $(command -v cygpath &> /dev/null) ; then
+    VIRTUAL_ENV=$(cygpath -u "$VIRTUAL_ENV")
+fi
+export VIRTUAL_ENV
+
+_OLD_VIRTUAL_PATH="$PATH"
+PATH="$VIRTUAL_ENV/Scripts:$PATH"
+export PATH
+
+if [ "x" != x ] ; then
+    VIRTUAL_ENV_PROMPT="() "
+else
+    VIRTUAL_ENV_PROMPT="($(basename "$VIRTUAL_ENV")) "
+fi
+export VIRTUAL_ENV_PROMPT
+
+# unset PYTHONHOME if set
+if ! [ -z "${PYTHONHOME+_}" ] ; then
+    _OLD_VIRTUAL_PYTHONHOME="$PYTHONHOME"
+    unset PYTHONHOME
+fi
+
+if [ -z "${VIRTUAL_ENV_DISABLE_PROMPT-}" ] ; then
+    _OLD_VIRTUAL_PS1="${PS1-}"
+    PS1="${VIRTUAL_ENV_PROMPT}${PS1-}"
+    export PS1
+fi
+
+# Make sure to unalias pydoc if it's already there
+alias pydoc 2>/dev/null >/dev/null && unalias pydoc || true
+
+pydoc () {
+    python -m pydoc "$@"
+}
+
+# The hash command must be called to get it to forget past
+# commands. Without forgetting past commands the $PATH changes
+# we made may not be respected
+hash -r 2>/dev/null

.venv/Scripts/activate.bat

@@ -0,0 +1,71 @@
+@REM Copyright (c) 2020-202x The virtualenv developers
+@REM
+@REM Permission is hereby granted, free of charge, to any person obtaining
+@REM a copy of this software and associated documentation files (the
+@REM "Software"), to deal in the Software without restriction, including
+@REM without limitation the rights to use, copy, modify, merge, publish,
+@REM distribute, sublicense, and/or sell copies of the Software, and to
+@REM permit persons to whom the Software is furnished to do so, subject to
+@REM the following conditions:
+@REM
+@REM The above copyright notice and this permission notice shall be
+@REM included in all copies or substantial portions of the Software.
+@REM
+@REM THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+@REM EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+@REM MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+@REM NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+@REM LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+@REM OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+@REM WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+@REM This file is UTF-8 encoded, so we need to update the current code page while executing it
+@for /f "tokens=2 delims=:." %%a in ('"%SystemRoot%\System32\chcp.com"') do @set _OLD_CODEPAGE=%%a
+@if defined _OLD_CODEPAGE (
+    @"%SystemRoot%\System32\chcp.com" 65001 > nul
+)
+
+@for %%i in ("C:\Users\Admin\Downloads\AIDevMinh\slm-profile-rag\.venv") do @set "VIRTUAL_ENV=%%~fi"
+
+@set "VIRTUAL_ENV_PROMPT="
+@if NOT DEFINED VIRTUAL_ENV_PROMPT (
+    @for %%d in ("%VIRTUAL_ENV%") do @set "VIRTUAL_ENV_PROMPT=%%~nxd"
+)
+
+@if defined _OLD_VIRTUAL_PROMPT (
+    @set "PROMPT=%_OLD_VIRTUAL_PROMPT%"
+) else (
+    @if not defined PROMPT (
+        @set "PROMPT=$P$G"
+    )
+    @if not defined VIRTUAL_ENV_DISABLE_PROMPT (
+        @set "_OLD_VIRTUAL_PROMPT=%PROMPT%"
+    )
+)
+@if not defined VIRTUAL_ENV_DISABLE_PROMPT (
+    @set "PROMPT=(%VIRTUAL_ENV_PROMPT%) %PROMPT%"
+)
+
+@REM Don't use () to avoid problems with them in %PATH%
+@if defined _OLD_VIRTUAL_PYTHONHOME @goto ENDIFVHOME
+    @set "_OLD_VIRTUAL_PYTHONHOME=%PYTHONHOME%"
+:ENDIFVHOME
+
+@set PYTHONHOME=
+
+@REM if defined _OLD_VIRTUAL_PATH (
+@if not defined _OLD_VIRTUAL_PATH @goto ENDIFVPATH1
+    @set "PATH=%_OLD_VIRTUAL_PATH%"
+:ENDIFVPATH1
+@REM ) else (
+@if defined _OLD_VIRTUAL_PATH @goto ENDIFVPATH2
+    @set "_OLD_VIRTUAL_PATH=%PATH%"
+:ENDIFVPATH2
+
+@set "PATH=%VIRTUAL_ENV%\Scripts;%PATH%"
+
+:END
+@if defined _OLD_CODEPAGE (
+    @"%SystemRoot%\System32\chcp.com" %_OLD_CODEPAGE% > nul
+    @set _OLD_CODEPAGE=
+)

.venv/Scripts/activate.csh

@@ -0,0 +1,76 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+# This file must be used with "source bin/activate.csh" *from csh*.
+# You cannot run it directly.
+# Created by Davide Di Blasi <davidedb@gmail.com>.
+
+set newline='\
+'
+
+alias deactivate 'test $?_OLD_VIRTUAL_PATH != 0 && setenv PATH "$_OLD_VIRTUAL_PATH:q" && unset _OLD_VIRTUAL_PATH; rehash; test $?_OLD_VIRTUAL_PROMPT != 0 && set prompt="$_OLD_VIRTUAL_PROMPT:q" && unset _OLD_VIRTUAL_PROMPT; unsetenv VIRTUAL_ENV; unsetenv VIRTUAL_ENV_PROMPT; test "\!:*" != "nondestructive" && unalias deactivate && unalias pydoc'
+
+# Unset irrelevant variables.
+deactivate nondestructive
+
+setenv VIRTUAL_ENV 'C:\Users\Admin\Downloads\AIDevMinh\slm-profile-rag\.venv'
+
+set _OLD_VIRTUAL_PATH="$PATH:q"
+setenv PATH "$VIRTUAL_ENV:q/Scripts:$PATH:q"
+
+
+
+if ('' != "") then
+    setenv VIRTUAL_ENV_PROMPT ''
+else
+    setenv VIRTUAL_ENV_PROMPT "$VIRTUAL_ENV:t:q"
+endif
+
+if ( $?VIRTUAL_ENV_DISABLE_PROMPT ) then
+    if ( $VIRTUAL_ENV_DISABLE_PROMPT == "" ) then
+        set do_prompt = "1"
+    else
+        set do_prompt = "0"
+    endif
+else
+    set do_prompt = "1"
+endif
+
+if ( $do_prompt == "1" ) then
+    # Could be in a non-interactive environment,
+    # in which case, $prompt is undefined and we wouldn't
+    # care about the prompt anyway.
+    if ( $?prompt ) then
+        set _OLD_VIRTUAL_PROMPT="$prompt:q"
+        if ( "$prompt:q" =~ *"$newline:q"* ) then
+            :
+        else
+            set prompt = '('"$VIRTUAL_ENV_PROMPT:q"') '"$prompt:q"
+        endif
+    endif
+endif
+
+unset env_name
+unset do_prompt
+
+alias pydoc python -m pydoc
+
+rehash

.venv/Scripts/activate.fish

@@ -0,0 +1,124 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+# This file must be used using `source bin/activate.fish` *within a running fish ( http://fishshell.com ) session*.
+# Do not run it directly.
+
+function _bashify_path -d "Converts a fish path to something bash can recognize"
+    set fishy_path $argv
+    set bashy_path $fishy_path[1]
+    for path_part in $fishy_path[2..-1]
+        set bashy_path "$bashy_path:$path_part"
+    end
+    echo $bashy_path
+end
+
+function _fishify_path -d "Converts a bash path to something fish can recognize"
+    echo $argv | tr ':' '\n'
+end
+
+function deactivate -d 'Exit virtualenv mode and return to the normal environment.'
+    # reset old environment variables
+    if test -n "$_OLD_VIRTUAL_PATH"
+        # https://github.com/fish-shell/fish-shell/issues/436 altered PATH handling
+        if test (echo $FISH_VERSION | head -c 1) -lt 3
+            set -gx PATH (_fishify_path "$_OLD_VIRTUAL_PATH")
+        else
+            set -gx PATH $_OLD_VIRTUAL_PATH
+        end
+        set -e _OLD_VIRTUAL_PATH
+    end
+
+    if test -n "$_OLD_VIRTUAL_PYTHONHOME"
+        set -gx PYTHONHOME "$_OLD_VIRTUAL_PYTHONHOME"
+        set -e _OLD_VIRTUAL_PYTHONHOME
+    end
+
+    if test -n "$_OLD_FISH_PROMPT_OVERRIDE"
+       and functions -q _old_fish_prompt
+        # Set an empty local `$fish_function_path` to allow the removal of `fish_prompt` using `functions -e`.
+        set -l fish_function_path
+
+        # Erase virtualenv's `fish_prompt` and restore the original.
+        functions -e fish_prompt
+        functions -c _old_fish_prompt fish_prompt
+        functions -e _old_fish_prompt
+        set -e _OLD_FISH_PROMPT_OVERRIDE
+    end
+
+    set -e VIRTUAL_ENV
+    set -e VIRTUAL_ENV_PROMPT
+
+    if test "$argv[1]" != 'nondestructive'
+        # Self-destruct!
+        functions -e pydoc
+        functions -e deactivate
+        functions -e _bashify_path
+        functions -e _fishify_path
+    end
+end
+
+# Unset irrelevant variables.
+deactivate nondestructive
+
+set -gx VIRTUAL_ENV 'C:\Users\Admin\Downloads\AIDevMinh\slm-profile-rag\.venv'
+
+# https://github.com/fish-shell/fish-shell/issues/436 altered PATH handling
+if test (echo $FISH_VERSION | head -c 1) -lt 3
+    set -gx _OLD_VIRTUAL_PATH (_bashify_path $PATH)
+else
+    set -gx _OLD_VIRTUAL_PATH $PATH
+end
+set -gx PATH "$VIRTUAL_ENV"'/Scripts' $PATH
+
+# Prompt override provided?
+# If not, just use the environment name.
+if test -n ''
+    set -gx VIRTUAL_ENV_PROMPT ''
+else
+    set -gx VIRTUAL_ENV_PROMPT (basename "$VIRTUAL_ENV")
+end
+
+# Unset `$PYTHONHOME` if set.
+if set -q PYTHONHOME
+    set -gx _OLD_VIRTUAL_PYTHONHOME $PYTHONHOME
+    set -e PYTHONHOME
+end
+
+function pydoc
+    python -m pydoc $argv
+end
+
+if test -z "$VIRTUAL_ENV_DISABLE_PROMPT"
+    # Copy the current `fish_prompt` function as `_old_fish_prompt`.
+    functions -c fish_prompt _old_fish_prompt
+
+    function fish_prompt
+        # Run the user's prompt first; it might depend on (pipe)status.
+        set -l prompt (_old_fish_prompt)
+
+        printf '(%s) ' $VIRTUAL_ENV_PROMPT
+
+        string join -- \n $prompt # handle multi-line prompts
+    end
+
+    set -gx _OLD_FISH_PROMPT_OVERRIDE "$VIRTUAL_ENV"
+end

.venv/Scripts/activate.nu

@@ -0,0 +1,117 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+# virtualenv activation module
+# Activate with `overlay use activate.nu`
+# Deactivate with `deactivate`, as usual
+#
+# To customize the overlay name, you can call `overlay use activate.nu as foo`,
+# but then simply `deactivate` won't work because it is just an alias to hide
+# the "activate" overlay. You'd need to call `overlay hide foo` manually.
+
+export-env {
+    def is-string [x] {
+        ($x | describe) == 'string'
+    }
+
+    def has-env [...names] {
+        $names | each {|n|
+            $n in $env
+        } | all {|i| $i == true}
+    }
+
+    # Emulates a `test -z`, but better as it handles e.g 'false'
+    def is-env-true [name: string] {
+      if (has-env $name) {
+        # Try to parse 'true', '0', '1', and fail if not convertible
+        let parsed = (do -i { $env | get $name | into bool })
+        if ($parsed | describe) == 'bool' {
+          $parsed
+        } else {
+          not ($env | get -i $name | is-empty)
+        }
+      } else {
+        false
+      }
+    }
+
+    let virtual_env = 'C:\Users\Admin\Downloads\AIDevMinh\slm-profile-rag\.venv'
+    let bin = 'Scripts'
+
+    let is_windows = ($nu.os-info.family) == 'windows'
+    let path_name = (if (has-env 'Path') {
+            'Path'
+        } else {
+            'PATH'
+        }
+    )
+
+    let venv_path = ([$virtual_env $bin] | path join)
+    let new_path = ($env | get $path_name | prepend $venv_path)
+
+    # If there is no default prompt, then use the env name instead
+    let virtual_env_prompt = (if ('' | is-empty) {
+        ($virtual_env | path basename)
+    } else {
+        ''
+    })
+
+    let new_env = {
+        $path_name         : $new_path
+        VIRTUAL_ENV        : $virtual_env
+        VIRTUAL_ENV_PROMPT : $virtual_env_prompt
+    }
+
+    let new_env = (if (is-env-true 'VIRTUAL_ENV_DISABLE_PROMPT') {
+      $new_env
+    } else {
+      # Creating the new prompt for the session
+      let virtual_prefix = $'(char lparen)($virtual_env_prompt)(char rparen) '
+
+      # Back up the old prompt builder
+      let old_prompt_command = (if (has-env 'PROMPT_COMMAND') {
+              $env.PROMPT_COMMAND
+          } else {
+              ''
+        })
+
+      let new_prompt = (if (has-env 'PROMPT_COMMAND') {
+          if 'closure' in ($old_prompt_command | describe) {
+              {|| $'($virtual_prefix)(do $old_prompt_command)' }
+          } else {
+              {|| $'($virtual_prefix)($old_prompt_command)' }
+          }
+      } else {
+          {|| $'($virtual_prefix)' }
+      })
+
+      $new_env | merge {
+        PROMPT_COMMAND      : $new_prompt
+        VIRTUAL_PREFIX      : $virtual_prefix
+      }
+    })
+
+    # Environment variables that will be loaded as the virtual env
+    load-env $new_env
+}
+
+export alias pydoc = python -m pydoc
+export alias deactivate = overlay hide activate

.venv/Scripts/activate.ps1

@@ -0,0 +1,82 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+$script:THIS_PATH = $myinvocation.mycommand.path
+$script:BASE_DIR = Split-Path (Resolve-Path "$THIS_PATH/..") -Parent
+
+function global:deactivate([switch] $NonDestructive) {
+    if (Test-Path variable:_OLD_VIRTUAL_PATH) {
+        $env:PATH = $variable:_OLD_VIRTUAL_PATH
+        Remove-Variable "_OLD_VIRTUAL_PATH" -Scope global
+    }
+
+    if (Test-Path function:_old_virtual_prompt) {
+        $function:prompt = $function:_old_virtual_prompt
+        Remove-Item function:\_old_virtual_prompt
+    }
+
+    if ($env:VIRTUAL_ENV) {
+        Remove-Item env:VIRTUAL_ENV -ErrorAction SilentlyContinue
+    }
+
+    if ($env:VIRTUAL_ENV_PROMPT) {
+        Remove-Item env:VIRTUAL_ENV_PROMPT -ErrorAction SilentlyContinue
+    }
+
+    if (!$NonDestructive) {
+        # Self destruct!
+        Remove-Item function:deactivate
+        Remove-Item function:pydoc
+    }
+}
+
+function global:pydoc {
+    python -m pydoc $args
+}
+
+# unset irrelevant variables
+deactivate -nondestructive
+
+$VIRTUAL_ENV = $BASE_DIR
+$env:VIRTUAL_ENV = $VIRTUAL_ENV
+
+if ("" -ne "") {
+    $env:VIRTUAL_ENV_PROMPT = ""
+}
+else {
+    $env:VIRTUAL_ENV_PROMPT = $( Split-Path $env:VIRTUAL_ENV -Leaf )
+}
+
+New-Variable -Scope global -Name _OLD_VIRTUAL_PATH -Value $env:PATH
+
+$env:PATH = "$env:VIRTUAL_ENV/Scripts;" + $env:PATH
+if (!$env:VIRTUAL_ENV_DISABLE_PROMPT) {
+    function global:_old_virtual_prompt {
+        ""
+    }
+    $function:_old_virtual_prompt = $function:prompt
+
+    function global:prompt {
+        # Add the custom prefix to the existing prompt
+        $previous_prompt_value = & $function:_old_virtual_prompt
+        ("(" + $env:VIRTUAL_ENV_PROMPT + ") " + $previous_prompt_value)
+    }
+}

.venv/Scripts/activate_this.py

@@ -0,0 +1,59 @@
+# Copyright (c) 2020-202x The virtualenv developers
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+"""
+Activate virtualenv for current interpreter:
+
+import runpy
+runpy.run_path(this_file)
+
+This can be used when you must use an existing Python interpreter, not the virtualenv bin/python.
+"""  # noqa: D415
+
+from __future__ import annotations
+
+import os
+import site
+import sys
+
+try:
+    abs_file = os.path.abspath(__file__)
+except NameError as exc:
+    msg = "You must use import runpy; runpy.run_path(this_file)"
+    raise AssertionError(msg) from exc
+
+bin_dir = os.path.dirname(abs_file)
+base = bin_dir[: -len("Scripts") - 1]  # strip away the bin part from the __file__, plus the path separator
+
+# prepend bin to PATH (this file is inside the bin directory)
+os.environ["PATH"] = os.pathsep.join([bin_dir, *os.environ.get("PATH", "").split(os.pathsep)])
+os.environ["VIRTUAL_ENV"] = base  # virtual env is right above bin directory
+os.environ["VIRTUAL_ENV_PROMPT"] = "" or os.path.basename(base)  # noqa: SIM222
+
+# add the virtual environments libraries to the host python import mechanism
+prev_length = len(sys.path)
+for lib in "..\\Lib\\site-packages".split(os.pathsep):
+    path = os.path.realpath(os.path.join(bin_dir, lib))
+    site.addsitedir(path)
+sys.path[:] = sys.path[prev_length:] + sys.path[0:prev_length]
+
+sys.real_prefix = sys.prefix
+sys.prefix = base

.venv/Scripts/deactivate.bat

@@ -0,0 +1,39 @@
+@REM Copyright (c) 2020-202x The virtualenv developers
+@REM
+@REM Permission is hereby granted, free of charge, to any person obtaining
+@REM a copy of this software and associated documentation files (the
+@REM "Software"), to deal in the Software without restriction, including
+@REM without limitation the rights to use, copy, modify, merge, publish,
+@REM distribute, sublicense, and/or sell copies of the Software, and to
+@REM permit persons to whom the Software is furnished to do so, subject to
+@REM the following conditions:
+@REM
+@REM The above copyright notice and this permission notice shall be
+@REM included in all copies or substantial portions of the Software.
+@REM
+@REM THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+@REM EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+@REM MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+@REM NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+@REM LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+@REM OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+@REM WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+@set VIRTUAL_ENV=
+@set VIRTUAL_ENV_PROMPT=
+
+@REM Don't use () to avoid problems with them in %PATH%
+@if not defined _OLD_VIRTUAL_PROMPT @goto ENDIFVPROMPT
+    @set "PROMPT=%_OLD_VIRTUAL_PROMPT%"
+    @set _OLD_VIRTUAL_PROMPT=
+:ENDIFVPROMPT
+
+@if not defined _OLD_VIRTUAL_PYTHONHOME @goto ENDIFVHOME
+    @set "PYTHONHOME=%_OLD_VIRTUAL_PYTHONHOME%"
+    @set _OLD_VIRTUAL_PYTHONHOME=
+:ENDIFVHOME
+
+@if not defined _OLD_VIRTUAL_PATH @goto ENDIFVPATH
+    @set "PATH=%_OLD_VIRTUAL_PATH%"
+    @set _OLD_VIRTUAL_PATH=
+:ENDIFVPATH

.venv/Scripts/pydoc.bat

@@ -0,0 +1,22 @@
+@REM Copyright (c) 2020-202x The virtualenv developers
+@REM
+@REM Permission is hereby granted, free of charge, to any person obtaining
+@REM a copy of this software and associated documentation files (the
+@REM "Software"), to deal in the Software without restriction, including
+@REM without limitation the rights to use, copy, modify, merge, publish,
+@REM distribute, sublicense, and/or sell copies of the Software, and to
+@REM permit persons to whom the Software is furnished to do so, subject to
+@REM the following conditions:
+@REM
+@REM The above copyright notice and this permission notice shall be
+@REM included in all copies or substantial portions of the Software.
+@REM
+@REM THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+@REM EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+@REM MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+@REM NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+@REM LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+@REM OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+@REM WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+python.exe -m pydoc %*

.venv/Scripts/python.exe

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b86e1ee66bac3fe3b998a2586791ef3151d5b7fee9af11f5b1491d2a4a93ff0
+size 237568

.venv/Scripts/pythonw.exe

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faaa4c9a1c0188a7bb7d648874b68c536db0248ca97b5d863854091864b94b3b
+size 233984

.venv/pyvenv.cfg

@@ -0,0 +1,5 @@
+home = C:\Users\Admin\AppData\Roaming\uv\python\cpython-3.13.2-windows-x86_64-none
+implementation = CPython
+uv = 0.6.9
+version_info = 3.13.2
+include-system-site-packages = false

ARCHITECTURE.md

@@ -0,0 +1,815 @@
+# 🏗️ Architecture Overview
+
+## System Architecture
+
+<p align="center">
+  <img src="docs/arch_system_design.png" alt="System Architecture Diagram" width="900" />
+</p>
+<p align="center"><em>High-level system architecture diagram illustrating the complete RAG pipeline flow</em></p>
+
+### Detailed Component View
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│                      User Interface                          │
+│                    (Streamlit Web App)                       │
+│                        app.py                                │
+└───────────────────────┬─────────────────────────────────────┘
+                        │
+                        ▼
+┌─────────────────────────────────────────────────────────────┐
+│                    RAG Pipeline                              │
+│                  (rag_pipeline.py)                           │
+│                                                              │
+│  ┌────────────────┐      ┌──────────────┐                  │
+│  │  System Prompt │      │  QA Chain    │                  │
+│  │   Template     │─────▶│  (LangChain) │                  │
+│  └────────────────┘      └──────┬───────┘                  │
+└─────────────────────────────────┼──────────────────────────┘
+                                  │
+                    ┌─────────────┴─────────────┐
+                    ▼                           ▼
+        ┌───────────────────────┐   ┌──────────────────────┐
+        │   Vector Store        │   │    LLM Handler       │
+        │  (vectorstore.py)     │   │  (llm_handler.py)    │
+        │                       │   │                      │
+        │  ┌─────────────────┐  │   │  ┌───────────────┐  │
+        │  │   ChromaDB      │  │   │  │    Ollama     │  │
+        │  │  (Embeddings)   │  │   │  │  (llama3.2)   │  │
+        │  └─────────────────┘  │   │  └───────────────┘  │
+        │                       │   │                      │
+        │  ┌─────────────────┐  │   └──────────────────────┘
+        │  │ HuggingFace     │  │
+        │  │ Embeddings      │  │
+        │  │ (all-MiniLM)    │  │
+        │  └─────────────────┘  │
+        └───────────────────────┘
+                    ▲
+                    │
+        ┌───────────┴────────────┐
+        │ Document Processor     │
+        │ (document_processor.py)│
+        │                        │
+        │  ┌──────────────────┐  │
+        │  │  PDF Loader      │  │
+        │  │  DOCX Loader     │  │
+        │  │  HTML Loader     │  │
+        │  │  Text Loader     │  │
+        │  └──────────────────┘  │
+        │                        │
+        │  ┌──────────────────┐  │
+        │  │ Text Splitter    │  │
+        │  │ (Chunking)       │  │
+        │  └──────────────────┘  │
+        └────────────────────────┘
+                    ▲
+                    │
+        ┌───────────┴────────────┐
+        │   Source Documents     │
+        │   data/documents/      │
+        │                        │
+        │  • Resume.pdf          │
+        │  • LinkedIn.html       │
+        │  • Projects.docx       │
+        │  • ...                 │
+        └────────────────────────┘
+
+┌──────────────────────────────────────────────────────────────┐
+│                 Configuration Layer                          │
+├──────────────────────────────────────────────────────────────┤
+│                                                              │
+│  config.yaml          env.template (.env)                   │
+│  ├─ profile          ├─ OLLAMA_BASE_URL                     │
+│  ├─ llm              ├─ CHROMA_PERSIST_DIR                  │
+│  ├─ embeddings       ├─ DOCUMENTS_DIR                       │
+│  ├─ vectorstore      ├─ LOG_LEVEL                           │
+│  ├─ document_proc    └─ API_KEYS (optional)                 │
+│  ├─ rag                                                      │
+│  ├─ ui                                                       │
+│  └─ logging                                                  │
+│                                                              │
+│         (config_loader.py)                                   │
+└──────────────────────────────────────────────────────────────┘
+```
+
+## Data Flow
+
+### 1️⃣ Indexing Phase (One-time Setup)
+
+```
+Documents → Load → Chunk → Build Indexes
+   ↓          ↓       ↓           ↓
+resume.pdf  PyPDF  Split    ┌─────────────────────────────┐
+linkedin.html BS4   into    │  BM25 Index (keyword)       │
+projects.docx docx  chunks  │  ./bm25_index/              │
+                            ├─────────────────────────────┤
+                            │  Vector Index (semantic)    │
+                            │  ./chroma_db/ (ChromaDB)    │
+                            └─────────────────────────────┘
+```
+
+**Command:** `python -m src.build_vectorstore`
+
+**Strategy Options:**
+- `--strategy bm25_vector` - Build both indexes (default, recommended)
+- `--strategy vector` - Vector index only
+- `--strategy bm25` - BM25 index only
+
+### 2️⃣ Query Phase (Runtime) - WITH HYBRID SEARCH
+
+```
+User Question
+     ↓
+┌─────────────────────────────────────┐
+│ Extract Chat History (if enabled)     │
+│  • Format previous Q&A pairs         │
+│  • Truncate by turns/tokens          │
+│  • Include in context                │
+└────────────┬────────────────────────┘
+             ↓
+┌─────────────────────────────────────┐
+│ Load Main Document (if enabled)      │
+│  • Check cache validity              │
+│  • Auto-detect format                │
+│  • Count tokens                      │
+│  • Summarize if needed               │
+└────────────┬────────────────────────┘
+             ↓
+      [Main Doc Content]
+             │
+             ├─────────────────────────────────────┐
+             ↓                                     ↓
+┌─────────────────────────────────────┐    Main Doc (Priority)
+│      Hybrid Retrieval Strategy       │          ↓
+│                                      │   [Always Available]
+│  ┌─────────────┐  ┌─────────────┐   │   [10k tokens max]
+│  │   BM25      │  │   Vector    │   │
+│  │  (keyword)  │  │ (semantic)  │   │
+│  └──────┬──────┘  └──────┬──────┘   │
+│         │                │          │
+│         └───────┬────────┘          │
+│                 ↓                   │
+│    Reciprocal Rank Fusion (RRF)     │
+│         (70% vector, 30% BM25)      │
+└────────────────┬────────────────────┘
+                 ↓
+         [Retrieved Context]
+                 │
+                 └─────────┬───────────────────┘
+                           ↓
+                  Construct Prompt:
+    System Prompt → Chat History → Main Doc → Retrieved Context → Question
+                           ↓
+                  Send to LLM (Ollama/Groq)
+                           ↓
+                  Generate Answer (streaming)
+                           ↓
+                  Response Enhancer
+                  • Detect negative phrases
+                  • Rewrite positively
+                  • Fix markdown formatting
+                  • Add forward-looking closings
+                           ↓
+             Display in Streamlit UI (with source citations)
+```
+
+## Component Details
+
+### 📑 Main Document Integration
+
+The Main Document feature ensures critical profile information is always available in the LLM context, regardless of VectorDB retrieval quality.
+
+```python
+MainDocumentLoader
+├── Format Auto-Detection
+│   ├── Markdown (.md)    → LangChain TextLoader
+│   ├── Plain Text (.txt) → LangChain TextLoader
+│   ├── PDF (.pdf)        → Existing PDF loader
+│   ├── Word (.docx)      → Existing DOCX loader
+│   └── HTML (.html)      → Existing HTML loader
+│
+├── Token Management
+│   ├── Counting: tiktoken (cl100k_base encoding)
+│   ├── Max Limit: 10,000 tokens (configurable)
+│   ├── Truncation: Smart token-based trimming
+│   └── Summarization: LLM-based if exceeds limit
+│
+├── Caching Strategy
+│   ├── File hash-based invalidation (MD5)
+│   ├── Configurable check interval (60s default)
+│   └── Automatic reload on file changes
+│
+└── Integration Point
+    └── Positioned BEFORE VectorDB context (high priority)
+```
+
+**Architecture Flow:**
+```
+Main Document (Priority Context)
+         ↓
+    [Essential Info Always Available]
+         ↓
+VectorDB Retrieval (Additional Context)
+         ↓
+    [Supplementary Information]
+         ↓
+Combined Context → LLM → Response
+```
+
+**Benefits:**
+- ✅ Critical information never missed by retrieval
+- ✅ Auto-format detection (no manual config)
+- ✅ Intelligent token management with LLM summarization
+- ✅ Efficient caching for performance
+- ✅ Graceful degradation if unavailable
+
+**Note:** The system also supports chat history for conversational context. Previous Q&A pairs are included in the prompt (after system prompt, before main document) with configurable truncation by turns or tokens. Responses are post-processed to improve tone, remove negative language, and fix markdown formatting.
+
+### 📄 Document Processing Pipeline
+
+```python
+DocumentProcessor
+├── Supported Formats
+│   ├── PDF          → pypdf
+│   ├── Word         → python-docx
+│   ├── HTML         → BeautifulSoup4
+│   └── Text/MD      → LangChain TextLoader
+│
+├── Chunking Strategy
+│   ├── Size: 1000 chars (configurable)
+│   ├── Overlap: 200 chars (configurable)
+│   └── Separators: ["\n\n", "\n", ". ", " ", ""]
+│
+└── Output: List[Document]
+    └── Each with content + metadata
+```
+
+### 🔍 Retrieval Strategy System
+
+The retrieval system uses a pluggable strategy pattern for extensibility.
+
+```python
+RetrieverFactory
+├── Registered Strategies
+│   ├── "vector"      → VectorStrategy (semantic search)
+│   ├── "bm25"        → BM25Strategy (keyword search)
+│   ├── "bm25_vector" → BM25VectorStrategy (hybrid)
+│   └── (future: "page_index", "graph_vector")
+│
+└── Strategy Interface (BaseRetrieverStrategy)
+    ├── build_index(documents) → Build/update index
+    ├── load_index()           → Load from disk
+    ├── retrieve(query, k)     → Get relevant docs
+    └── as_retriever()         → LangChain compatible
+```
+
+#### Hybrid Search (BM25 + Vector)
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                   BM25VectorStrategy                            │
+│                                                                 │
+│  ┌─────────────────────┐       ┌─────────────────────────┐     │
+│  │   BM25Retriever     │       │   VectorRetriever       │     │
+│  │   (keyword match)   │       │   (semantic match)      │     │
+│  │                     │       │                         │     │
+│  │  • Exact terms      │       │  • Meaning/context      │     │
+│  │  • Abbreviations    │       │  • Synonyms             │     │
+│  │  • Proper nouns     │       │  • Related concepts     │     │
+│  └──────────┬──────────┘       └───────────┬─────────────┘     │
+│             │ (k=10)                       │ (k=10)            │
+│             └───────���──┬───────────────────┘                   │
+│                        ↓                                       │
+│         Reciprocal Rank Fusion (RRF)                           │
+│         weights: {vector: 0.7, bm25: 0.3}                      │
+│                        ↓                                       │
+│              Top K results (k=4)                               │
+└─────────────────────────────────────────────────────────────────┘
+```
+
+**RRF Formula:** `score(d) = Σ (weight × 1/(k + rank(d)))`
+
+### 🧠 Vector Store Architecture
+
+```python
+VectorStoreManager
+├── Embedding Model
+│   └── sentence-transformers/all-MiniLM-L6-v2
+│       ├── Dimension: 384
+│       ├── Speed: ~2000 sentences/sec (CPU)
+│       └── Quality: Good for semantic search
+│
+├── ChromaDB
+│   ├── Type: Persistent (SQLite)
+│   ├── Location: ./chroma_db/
+│   ├── Collection: profile_documents
+│   └── Indexing: HNSW (approximate NN)
+│
+└── Retrieval
+    ├── Search Type: Similarity (or MMR)
+    ├── Top K: 10 (before fusion)
+    └── Distance: Cosine similarity
+```
+
+### 📝 BM25 Store Architecture
+
+```python
+BM25Store
+├── Algorithm: BM25Okapi (rank-bm25)
+│
+├── Tokenization
+│   ├── "simple": Whitespace + punctuation split
+│   └── "nltk": NLTK word_tokenize (optional)
+│
+├── Persistence
+│   ├── Location: ./bm25_index/
+│   ├── Format: Pickle (index + documents)
+│   └── Metadata: JSON (hash, stats)
+│
+└── Retrieval
+    ├── Top K: 10 (before fusion)
+    └── Scoring: BM25 term frequency
+```
+
+### 🤖 LLM Integration
+
+```python
+LLMHandler
+├── Provider Support
+│   ├── Ollama (Local)
+│   │   ├── Base URL: http://localhost:11434
+│   │   ├── Protocol: HTTP/REST API
+│   │   └── Auto-detected if no API key
+│   │
+│   └── Groq (Cloud)
+│       ├── Base URL: https://api.groq.com/openai/v1
+│       ├── Protocol: OpenAI-compatible API
+│       └── Requires GROQ_API_KEY env var
+│
+├── Model Options (Ollama)
+│   ├── llama3.2:3b  (Recommended)
+│   ├── phi3:mini
+│   ├── gemma2:2b
+│   └── llama3.1:8b  (with GPU)
+│
+├── Model Options (Groq)
+│   ├── openai/gpt-oss-120b
+│   ├── llama-3.3-70b-versatile
+│   ├── llama-3.1-8b-instant
+│   ├── mixtral-8x7b-32768
+│   └── gemma2-9b-it
+│
+└── Parameters
+    ├── Temperature: 0.7
+    ├── Max Tokens: 800 (increased for better formatting)
+    ├── Top P: 0.9
+    └── Context Window: Model-dependent (8192-128k tokens)
+```
+
+### 🔗 RAG Chain
+
+```python
+RAGPipeline
+├── Retrieval Strategy
+│   └── RetrieverFactory.create(strategy_name)
+│       ├── "vector"      → Vector-only retriever
+│       ├── "bm25"        → BM25-only retriever
+│       └── "bm25_vector" → Fusion retriever (default)
+│
+├── Prompt Template Structure
+│   ├── System Prompt (from config)
+│   ├── Chat History (formatted previous Q&A pairs)
+│   ├── Main Document (priority context, always available)
+│   ├── Retrieved Context (from strategy)
+│   └── User Question
+│
+├── LLM
+│   └── Ollama or Groq (auto-detected or configured)
+│
+└── Output
+    ├── Answer (post-processed for tone/formatting, streaming support)
+    └── Source Documents (citations)
+```
+
+## Configuration Hierarchy
+
+```
+1. Environment Variables (.env)
+   ├── Override config.yaml values
+   ├── Secrets (API keys)
+   └── Runtime settings (ports, URLs)
+      ↓
+2. config.yaml
+   ├── Application defaults
+   ├── Model selection
+   ├── Retrieval strategy (vector, bm25, bm25_vector)
+   └── RAG parameters
+      ↓
+3. Code Defaults
+   └── Fallback values if config missing
+```
+
+### Retrieval Configuration
+
+```yaml
+retrieval:
+  strategy: "bm25_vector"      # Which strategy to use
+  final_k: 4                   # Documents returned to LLM
+
+  vector:
+    search_type: "similarity"  # or "mmr"
+    k: 10                      # Docs before fusion
+
+  bm25:
+    k: 10                      # Docs before fusion
+    persist_path: "./bm25_index"
+    tokenizer: "simple"
+
+  fusion:
+    algorithm: "rrf"           # Reciprocal Rank Fusion
+    rrf_k: 60                  # RRF constant
+    weights:
+      vector: 0.7
+      bm25: 0.3
+```
+
+### Chat History Configuration
+
+```yaml
+chat:
+  enable_history: true         # Enable conversation context
+  max_history_turns: 10        # Max Q&A pairs to include
+  max_history_tokens: 2000      # Token limit for history
+```
+
+### RAG Configuration
+
+```yaml
+rag:
+  enhance_responses: true       # Enable post-processing enhancement
+  include_sources: true         # Show source citations
+  source_max_length: 150        # Max length of source preview
+```
+
+## Deployment Architecture
+
+### Local Development
+
+```
+┌──────────────────┐
+│   Developer      │
+│   Machine        │
+│                  │
+│  ┌────────────┐  │
+│  │  Ollama    │  │  ← Port 11434 (optional)
+│  │  Server    │  │
+│  └────────────┘  │
+│                  │
+│  ┌────────────┐  │
+│  │  Groq API  │  │  ← https://api.groq.com (optional)
+│  │  (Cloud)   │  │
+│  └────────────┘  │
+│                  │
+│  ┌────────────┐  │
+│  │ Streamlit  │  │  ← Port 8501
+│  │    App     │  │
+│  └────────────┘  │
+│                  │
+│  ┌────────────┐  │
+│  │ ChromaDB   │  │  ← ./chroma_db/
+│  │  (local)   │  │
+│  └────────────┘  │
+│                  │
+│  ┌────────────┐  │
+│  │ BM25 Index │  │  ← ./bm25_index/
+│  │  (local)   │  │
+│  └────────────┘  │
+└──────────────────┘
+```
+
+### Hugging Face Spaces
+
+```
+┌───────────────────────────────┐
+│   HF Spaces Container         │
+│                               │
+│  ┌─────────────────────────┐  │
+│  │  Dockerfile             │  │
+│  │  ├─ Install Ollama      │  │
+│  │  ├─ Pull Model          │  │
+│  │  ├─ Build Indexes       │  │
+│  │  └─ Start Services      │  │
+│  └─────────────────────────┘  │
+│                               │
+│  ┌──────────┐  ┌──────────┐  │
+│  │  Ollama  │  │Streamlit │  │
+│  │  Server  │  │   App    │  │
+│  └──────────┘  └──────────┘  │
+│                               │
+│  ┌─────────────────────────┐  │
+│  │  Persistent Storage     │  │
+│  │  ├─ chroma_db/          │  │
+│  │  └─ bm25_index/          │  │
+│  └─────────────────────────┘  │
+└───────────────────────────────┘
+         ↑
+         │ HTTPS
+         │
+┌────────┴──────────┐
+│   Public Users    │
+│   (Recruiters)    │
+└───────────────────┘
+```
+
+## Build & Packaging
+
+### UV + Hatchling + Versioningit
+
+```
+pyproject.toml
+├── [build-system]
+│   ├── requires: ["hatchling", "versioningit"]
+│   └── build-backend: "hatchling.build"
+│
+├── [project]
+│   ├── name: "profillybot"
+│   ├── version: <from git tags via versioningit>
+│   └── dependencies: [...]
+│
+├── [tool.versioningit]
+│   ├── Read git tags (v0.1.0, v0.2.0, etc.)
+│   ├── Generate version string
+│   └── Write to src/_version.py
+│
+└── [tool.ruff]
+    ├── Linting rules
+    └── Formatting config
+```
+
+### Version from Git Tags
+
+```bash
+# Tag release
+git tag v0.1.0
+git push origin v0.1.0
+
+# Version automatically set
+python -c "from src import __version__; print(__version__)"
+# Output: 0.1.0
+
+# Development version (after tag)
+# Output: 0.1.0+5.g1a2b3c4  (5 commits after v0.1.0)
+```
+
+## Code Quality Pipeline
+
+```
+Developer Writes Code
+        ↓
+┌───────────────────┐
+│   Pre-commit      │
+│   (Optional)      │
+└─────────┬─────────┘
+          ↓
+┌───────────────────┐
+│  Ruff Check       │  ← Linting
+│  Ruff Format      │  ← Formatting
+└─────────┬─────────┘
+          ↓
+┌───────────────────┐
+│  Git Commit       │
+└─────────┬─────────┘
+          ↓
+┌───────────────────┐
+│  Push to GitHub   │
+└─────────┬─────────┘
+          ↓
+┌───────────────────┐
+│  GitHub Actions   │
+│  • Lint Check     │
+│  • Format Check   │
+│  • Tests (future) │
+└───────────────────┘
+```
+
+## Module Dependencies
+
+```
+app.py
+ ├─ src.config_loader
+ ├─ src.rag_pipeline
+ ├─ src.llm_handler (provider detection)
+ ├─ src.response_enhancer (post-stream enhancement)
+ ├─ src.main_document_loader (token counting for history)
+ └─ streamlit (UI, chat history management)
+
+src.rag_pipeline
+ ├─ src.config_loader
+ ├─ src.llm_handler
+ ├─ src.retrieval (RetrieverFactory, strategies)
+ ├─ src.main_document_loader
+ ├─ src.response_enhancer
+ └─ langchain (LCEL chains)
+
+src.retrieval
+ ├─ src.retrieval.base (BaseRetrieverStrategy)
+ ├─ src.retrieval.factory (RetrieverFactory)
+ ├─ src.retrieval.fusion (RRF, FusionRetriever)
+ ├─ src.retrieval.stores.bm25_store
+ └─ src.retrieval.strategies.*
+
+src.retrieval.strategies.vector
+ ├─ src.vectorstore
+ └─ src.retrieval.base
+
+src.retrieval.strategies.bm25
+ ├─ src.retrieval.stores.bm25_store
+ └─ rank_bm25
+
+src.retrieval.strategies.bm25_vector
+ ├─ src.retrieval.strategies.vector
+ ├─ src.retrieval.strategies.bm25
+ └─ src.retrieval.fusion
+
+src.main_document_loader
+ ├─ src.config_loader
+ ├─ src.document_processor
+ ├─ src.llm_handler
+ ├─ tiktoken
+ └─ pathlib, hashlib, time
+
+src.response_enhancer
+ ├─ src.config_loader
+ └─ re (regex pattern matching)
+
+src.vectorstore
+ ├─ src.config_loader
+ ├─ chromadb
+ └─ langchain_huggingface
+
+src.llm_handler
+ ├─ src.config_loader
+ ├─ langchain_community.llms (Ollama)
+ └─ langchain_groq (Groq support)
+
+src.document_processor
+ ├─ src.config_loader
+ ├─ pypdf
+ ├─ python-docx
+ ├─ beautifulsoup4
+ └─ langchain
+
+src.config_loader
+ ├─ pyyaml
+ └─ python-dotenv
+
+src.build_vectorstore
+ ├─ src.document_processor
+ ├─ src.retrieval (RetrieverFactory)
+ └─ src.vectorstore
+```
+
+## Performance Characteristics
+
+### Indexing (One-time)
+
+| Documents | Chunks | Embedding Time | ChromaDB Insert | Total |
+|-----------|--------|----------------|-----------------|-------|
+| 5 PDFs    | ~100   | ~5 seconds     | ~1 second       | ~6s   |
+| 20 PDFs   | ~400   | ~20 seconds    | ~2 seconds      | ~22s  |
+| 50 PDFs   | ~1000  | ~50 seconds    | ~5 seconds      | ~55s  |
+
+### Query (Runtime)
+
+| Step | Time (CPU) | Time (GPU) | Notes |
+|------|------------|------------|-------|
+| Chat history extraction | <1ms | <1ms | Token counting overhead |
+| Main doc loading | <1ms | <1ms | Cached after first load |
+| Embed query | 50ms | 10ms | Sentence transformers |
+| Vector search | 10-50ms | 10-50ms | ChromaDB HNSW |
+| BM25 search | 5-20ms | 5-20ms | In-memory index |
+| Fusion (RRF) | <1ms | <1ms | Rank combination |
+| LLM inference (Ollama) | 2-5s | 0.5-1s | Local model |
+| LLM inference (Groq) | 0.5-2s | N/A | Cloud API |
+| Response enhancement | <10ms | <10ms | Post-processing |
+| **Total (Ollama)** | **2-5s** | **0.5-1s** | |
+| **Total (Groq)** | **0.6-2s** | **N/A** | Faster cloud option |
+
+### Memory Usage
+
+| Component | RAM | Disk |
+|-----------|-----|------|
+| Streamlit | ~200MB | - |
+| Ollama (llama3.2:3b) | ~2GB | ~2GB |
+| ChromaDB | ~100MB | ~50MB per 1k docs |
+| BM25 Index | ~50MB | ~10MB per 1k docs |
+| Embeddings | ~500MB | ~500MB |
+| Chat History | ~1-10MB | - |
+| Response Enhancer | <1MB | - |
+| **Total** | **~3GB** | **~3GB** |
+
+## Security Architecture
+
+```
+User Input
+    ↓
+┌─────────────────────┐
+│ Input Validation    │  ← Length limits
+│ (Streamlit)         │  ← Character filtering
+└─────────┬───────────┘
+          ↓
+┌─────────────────────┐
+│ RAG Pipeline        │  ← Context isolation
+└─────────┬───────────┘
+          ↓
+┌─────────────────────┐
+│ LLM (Local)         │  ← No external API calls
+│                     │  ← Data stays local
+└─────────────────────┘
+
+Secrets Management:
+├─ .env (local)
+├─ .gitignore (.env excluded)
+└─ HF Spaces Secrets (cloud)
+```
+
+## Extensibility Points
+
+### 🔌 Plugin Architecture
+
+```python
+# Easy to extend:
+
+# 1. New document types
+DocumentProcessor.load_custom_format()
+
+# 2. New LLM providers
+LLMHandler.get_openai_llm()
+LLMHandler.get_anthropic_llm()
+# Already supports: Ollama (local), Groq (cloud)
+
+# 3. New retrieval strategies (extensible system!)
+@RetrieverFactory.register("page_index")
+class PageIndexStrategy(BaseRetrieverStrategy):
+    """Vision-based document retrieval (ColPali)"""
+    ...
+
+@RetrieverFactory.register("graph_vector")
+class GraphVectorStrategy(BaseRetrieverStrategy):
+    """Knowledge graph + vector hybrid"""
+    ...
+
+# 4. New UI features
+app.py → add_authentication()
+app.py → add_analytics()
+# Already supports: Chat history, provider switching, streaming responses
+
+# 5. New embedding models
+VectorStoreManager(embedding_model="...")
+
+# 6. New fusion algorithms
+# Add to src/retrieval/fusion.py
+def custom_fusion(results_list, weights):
+    ...
+
+# 7. New response enhancement patterns
+# Edit src/response_enhancer.py
+ResponseEnhancer.negative_patterns.append((pattern, rewrite_func))
+
+# 8. Custom chat history strategies
+# Modify truncate_chat_history() in app.py
+# Add custom truncation logic (e.g., importance-based)
+```
+
+### Adding a New Retrieval Strategy
+
+1. Create `src/retrieval/strategies/my_strategy.py`
+2. Implement `BaseRetrieverStrategy` interface
+3. Register with `@RetrieverFactory.register("my_strategy")`
+4. Add config section in `config.yaml`
+5. Import in `src/retrieval/strategies/__init__.py`
+
+```python
+from src.retrieval import RetrieverFactory
+from src.retrieval.base import BaseRetrieverStrategy
+
+@RetrieverFactory.register("my_strategy")
+class MyStrategy(BaseRetrieverStrategy):
+    @property
+    def name(self) -> str:
+        return "my_strategy"
+
+    def build_index(self, documents): ...
+    def load_index(self) -> bool: ...
+    def retrieve(self, query, k=4): ...
+    def as_retriever(self, **kwargs): ...
+```
+
+---
+
+**This architecture prioritizes:**
+- ✅ Simplicity (easy to understand)
+- ✅ Modularity (easy to extend)
+- ✅ Performance (optimized for small-medium datasets)
+- ✅ Privacy (local processing)
+- ✅ Deployability (cloud-ready)

Dockerfile

@@ -0,0 +1,68 @@
+# Dockerfile for HuggingFace Spaces with Ollama support
+FROM python:3.11-slim
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    curl \
+    build-essential \
+    zstd \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Ollama
+RUN curl -fsSL https://ollama.ai/install.sh | sh
+
+# Copy requirements
+COPY requirements.txt .
+
+# Install CPU-only PyTorch first (saves ~2GB by avoiding CUDA dependencies)
+RUN pip install --no-cache-dir torch --index-url https://download.pytorch.org/whl/cpu
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY . .
+
+# Create necessary directories
+RUN mkdir -p data/documents chroma_db bm25_index
+
+# Expose Streamlit port
+EXPOSE 7860
+
+# Expose Ollama port
+EXPOSE 11434
+
+# Start script that runs both Ollama and Streamlit
+COPY <<'EOF' /app/start.sh
+#!/bin/bash
+set -e
+
+# Start Ollama in the background
+ollama serve &
+OLLAMA_PID=$!
+
+# Wait for Ollama to be ready
+echo "Waiting for Ollama to start..."
+sleep 5
+
+# Pull the model (use small model for HF Spaces)
+MODEL=${OLLAMA_MODEL:-llama3.2:3b}
+echo "Pulling model: $MODEL"
+ollama pull $MODEL || echo "Model pull failed, will retry on app start"
+
+# Build retrieval indexes if documents exist and indexes don't
+if [ -d "data/documents" ] && [ ! -f "chroma_db/chroma.sqlite3" ]; then
+    echo "Building retrieval indexes (BM25 + Vector)..."
+    python -m src.build_vectorstore || echo "Index build failed, will build on first use"
+fi
+
+# Start Streamlit
+streamlit run app.py --server.port=7860 --server.address=0.0.0.0
+EOF
+
+RUN chmod +x /app/start.sh
+
+CMD ["/app/start.sh"]

Dockerfile.app

@@ -0,0 +1,80 @@
+# Dockerfile for Streamlit RAG Application (for Docker Compose)
+# Use with docker-compose.yml - Ollama runs as a separate service
+FROM python:3.11-slim
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    curl \
+    build-essential \
+    zstd \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements
+COPY requirements.txt .
+
+# Install CPU-only PyTorch first (saves ~2GB by avoiding CUDA dependencies)
+RUN pip install --no-cache-dir torch --index-url https://download.pytorch.org/whl/cpu
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY src/ ./src/
+COPY app.py .
+COPY config.yaml .
+COPY .streamlit/ ./.streamlit/
+
+# Create necessary directories
+RUN mkdir -p data/documents chroma_db bm25_index
+
+# Create startup script
+COPY <<'EOF' /app/start.sh
+#!/bin/bash
+set -e
+
+echo "Starting ProfillyBot Application..."
+
+# Wait for Ollama to be available (if using Ollama)
+if [ -n "$OLLAMA_HOST" ]; then
+    echo "Waiting for Ollama at $OLLAMA_HOST..."
+    for i in {1..30}; do
+        if curl -sf "$OLLAMA_HOST/api/tags" > /dev/null 2>&1; then
+            echo "Ollama is ready!"
+            break
+        fi
+        echo "Waiting for Ollama... ($i/30)"
+        sleep 2
+    done
+fi
+
+# Build retrieval indexes if documents exist and indexes don't
+if [ -d "data/documents" ] && [ "$(ls -A data/documents 2>/dev/null)" ]; then
+    if [ ! -f "chroma_db/chroma.sqlite3" ]; then
+        echo "Building retrieval indexes (BM25 + Vector)..."
+        python -m src.build_vectorstore || echo "Index build failed, will build on first use"
+    else
+        echo "Indexes already exist, skipping build"
+    fi
+fi
+
+# Start Streamlit
+echo "Starting Streamlit on port ${STREAMLIT_SERVER_PORT:-7860}..."
+exec streamlit run app.py \
+    --server.port=${STREAMLIT_SERVER_PORT:-7860} \
+    --server.address=${STREAMLIT_SERVER_ADDRESS:-0.0.0.0} \
+    --server.headless=true
+EOF
+
+RUN chmod +x /app/start.sh
+
+# Expose Streamlit port
+EXPOSE 7860
+
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=30s --retries=3 \
+    CMD curl -f http://localhost:7860/_stcore/health || exit 1
+
+CMD ["/app/start.sh"]

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Tin-Hoang
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Makefile

@@ -0,0 +1,146 @@
+# Makefile for ProfillyBot
+# Run 'make help' to see available commands
+
+.PHONY: help install install-dev test test-cov lint format clean \
+        docker-up docker-down docker-build docker-logs docker-shell \
+        docker-rebuild docker-ps ollama-pull index-build run
+
+# Default target
+.DEFAULT_GOAL := help
+
+# Variables
+PYTHON := python
+DOCKER_COMPOSE := docker compose
+APP_PORT := 7860
+OLLAMA_MODEL ?= llama3.2:3b
+
+# =============================================================================
+# Help
+# =============================================================================
+help: ## Show this help message
+	@echo "Usage: make [target]"
+	@echo ""
+	@echo "Available targets:"
+	@awk 'BEGIN {FS = ":.*##"; printf ""} /^[a-zA-Z_-]+:.*?##/ { printf "  \033[36m%-20s\033[0m %s\n", $$1, $$2 }' $(MAKEFILE_LIST)
+
+# =============================================================================
+# Development Setup
+# =============================================================================
+install: ## Install dependencies
+	pip install -r requirements.txt
+
+install-dev: ## Install development dependencies
+	pip install -e ".[dev]"
+
+install-uv: ## Install dependencies using uv (faster)
+	uv pip install -r requirements.txt
+
+install-uv-dev: ## Install dev dependencies using uv
+	uv pip install -e ".[dev]"
+
+# =============================================================================
+# Testing & Quality
+# =============================================================================
+test: ## Run tests
+	pytest tests/ -v
+
+test-cov: ## Run tests with coverage report
+	pytest tests/ -v --cov=src --cov-report=term-missing --cov-report=html
+
+lint: ## Run linter (ruff)
+	ruff check src/ tests/ app.py
+
+lint-fix: ## Run linter and fix issues
+	ruff check src/ tests/ app.py --fix
+
+format: ## Format code (ruff)
+	ruff format src/ tests/ app.py
+
+format-check: ## Check code formatting
+	ruff format src/ tests/ app.py --check
+
+check: lint format-check ## Run all checks (lint + format)
+
+# =============================================================================
+# Local Development
+# =============================================================================
+run: ## Run the Streamlit app locally
+	streamlit run app.py --server.port=$(APP_PORT)
+
+index-build: ## Build vector and BM25 indexes
+	$(PYTHON) -m src.build_vectorstore
+
+index-clean: ## Remove existing indexes
+	rm -rf chroma_db/ bm25_index/
+
+index-rebuild: index-clean index-build ## Rebuild indexes from scratch
+
+# =============================================================================
+# Docker Compose
+# =============================================================================
+docker-up: ## Start all services
+	$(DOCKER_COMPOSE) up -d
+
+docker-down: ## Stop all services
+	$(DOCKER_COMPOSE) down
+
+docker-build: ## Build Docker images
+	$(DOCKER_COMPOSE) build
+
+docker-rebuild: ## Rebuild and restart services
+	$(DOCKER_COMPOSE) up -d --build
+
+docker-logs: ## Show logs (follow mode)
+	$(DOCKER_COMPOSE) logs -f
+
+docker-logs-app: ## Show app logs only
+	$(DOCKER_COMPOSE) logs -f app
+
+docker-ps: ## Show running containers
+	$(DOCKER_COMPOSE) ps
+
+docker-shell: ## Open shell in app container
+	$(DOCKER_COMPOSE) exec app /bin/bash
+
+docker-clean: ## Stop services and remove volumes
+	$(DOCKER_COMPOSE) down -v
+
+docker-restart: docker-down docker-up ## Restart all services
+
+# =============================================================================
+# Ollama Commands
+# =============================================================================
+ollama-pull: ## Pull the configured Ollama model
+	$(DOCKER_COMPOSE) exec ollama ollama pull $(OLLAMA_MODEL)
+
+ollama-list: ## List available Ollama models
+	$(DOCKER_COMPOSE) exec ollama ollama list
+
+ollama-shell: ## Open shell in Ollama container
+	$(DOCKER_COMPOSE) exec ollama /bin/bash
+
+# =============================================================================
+# Setup & Utilities
+# =============================================================================
+setup: ## Initial project setup (copy env, install deps)
+	@if [ ! -f .env ]; then cp .env.template .env && echo "Created .env from template"; fi
+	@echo "Installing dependencies..."
+	@make install-dev
+	@echo "Setup complete! Edit .env if needed, then run 'make run' or 'make docker-up'"
+
+clean: ## Clean up build artifacts and caches
+	rm -rf __pycache__ .pytest_cache .ruff_cache htmlcov .coverage
+	find . -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
+	find . -type f -name "*.pyc" -delete 2>/dev/null || true
+
+clean-all: clean index-clean ## Clean everything including indexes
+	rm -rf *.egg-info build dist
+
+# =============================================================================
+# Pre-commit
+# =============================================================================
+pre-commit-install: ## Install pre-commit hooks
+	pre-commit install
+
+pre-commit-run: ## Run pre-commit on all files
+	pre-commit run --all-files

README.md

@@ -1,13 +1,423 @@
 ---
 title: ProfillyBot
-emoji: 🐨
-colorFrom: pink
-colorTo: yellow
+emoji: 🤖
+colorFrom: blue
+colorTo: green
 sdk: gradio
-sdk_version: 6.20.0
-python_version: '3.12'
-app_file: app.py
+sdk_version: 5.49.1
+python_version: "3.11"
+app_file: gradio_app.py
 pinned: false
+license: mit
+short_description: Cute RAG chatbot for professional profile Q&A on ZeroGPU
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🤖 ProfillyBot
+
+<p align="center">
+  <img src="docs/banner_image.png" alt="ProfillyBot Banner" width="100%" />
+</p>
+
+<p align="center">
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python310.yml/badge.svg" alt="Python 3.10"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python311.yml/badge.svg" alt="Python 3.11"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-python312.yml/badge.svg" alt="Python 3.12"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/test-docker.yml/badge.svg" alt="Docker"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/lint.yml/badge.svg" alt="Lint"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/code-quality.yml/badge.svg" alt="Code Quality"/>
+  <img src="https://github.com/Tin-Hoang/slm-profile-rag/actions/workflows/security-scan.yml/badge.svg" alt="Security Scan"/>
+  <a href="https://codecov.io/gh/Tin-Hoang/slm-profile-rag">
+    <img src="https://codecov.io/gh/Tin-Hoang/slm-profile-rag/graph/badge.svg?token=MW4RD9R66J" alt="Codecov"/>
+  </a>
+</p>
+
+**ProfillyBot** is a RAG (Retrieval Augmented Generation) chatbot that answers questions about your professional profile using your resume, project reports, and other documents.
+
+## ✨ Features
+
+- 📄 **Multi-format Support**: Process PDF, Word, HTML, and text documents
+- 🔍 **Hybrid Search**: Combines BM25 (keyword) + Vector (semantic) search with Reciprocal Rank Fusion
+- 🧠 **Extensible Retrieval**: Pluggable strategy system - easily add new retrieval methods (PageIndex, GraphRAG, etc.)
+- 📌 **Main Document Support**: Guaranteed context - critical information always available, auto-format detection
+- 🦙 **Ollama Integration**: Run small language models locally
+- 🎨 **Clean UI**: Streamlit-based interface
+- ⚙️ **Highly Configurable**: YAML-based settings for easy customization
+- 🚀 **HuggingFace Spaces Ready**: Deploy with one click
+- ✨ **Smart Response Enhancement**: Automatically removes negative language and adds professional, recruiter-friendly tone
+
+## 🏗️ Architecture
+
+<p align="center">
+  <img src="docs/arch_system_design.png" alt="System Architecture Diagram" width="800" />
+</p>
+<p align="center"><em>Overall system design showing the RAG pipeline and component interactions</em></p>
+
+Detailed architecture can be found in [ARCHITECTURE.md](ARCHITECTURE.md).
+
+## 🛠️ Tech Stack
+
+- **Python 3.10+** with UV package manager
+- **LangChain** for RAG pipeline
+- **ChromaDB** for vector storage
+- **rank-bm25** for lexical search
+- **Ollama** for local LLM serving
+- **Streamlit** for web interface
+- **sentence-transformers** for embeddings
+- **tiktoken** for token counting
+- **Ruff** for linting/formatting
+
+## 📦 Installation
+
+### Prerequisites
+
+1. **Python 3.10+**
+2. **UV Package Manager**: Install via:
+   ```bash
+   curl -LsSf https://astral.sh/uv/install.sh | sh
+   ```
+3. **Ollama**: Install from [ollama.ai](https://ollama.ai)
+
+### Setup
+
+1. **Clone the repository**:
+   ```bash
+   git clone <your-repo-url>
+   cd profillybot
+   ```
+
+2. **Install dependencies with UV**:
+   ```bash
+   uv venv
+   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+   uv pip install -r requirements.txt
+   ```
+
+3. **Set up environment variables**:
+   ```bash
+   cp .env.template .env
+   # Edit .env with your settings
+   ```
+
+4. **Configure the chatbot**:
+   - Edit `config.yaml` to set your name, title, and preferences
+   - Adjust model settings, chunking parameters, and system prompt
+
+5. **Add your documents**:
+   ```bash
+   # Place your PDFs, Word docs, HTML files in:
+   mkdir -p data/documents
+   # Copy your resume, project reports, LinkedIn profile, etc.
+   ```
+
+6. **Create your Main Profile Document** (📌 **Recommended**):
+
+   The Main Document feature ensures critical information is **always** included in responses, never missed by vector search.
+
+   ```bash
+   # Create main profile document (any format: .md, .txt, .pdf, .docx, .html)
+   nano data/documents/main_profile.md
+   ```
+
+   **Include essential information:**
+   - Full name, title, contact info
+   - Current role and key responsibilities
+   - Core skills and expertise
+   - Major projects with metrics
+   - Education and certifications
+
+   **Example structure:**
+   ```markdown
+   # Your Name
+   **Title**: Your Professional Title
+   **Email**: your.email@example.com
+
+   ## Professional Summary
+   Brief summary of your experience...
+
+   ## Core Skills
+   - Skill 1, Skill 2, Skill 3
+
+   ## Current Role
+   ### Company - Role (Dates)
+   - Achievement 1 with metrics
+   - Achievement 2 with metrics
+
+   ## Education
+   - Degree, University, Year
+   ```
+
+   **Why use this?**
+   - ✅ **Guaranteed Context**: Critical info never missed by vector similarity search
+   - ✅ **Priority Positioning**: Appears BEFORE retrieved chunks (higher LLM attention)
+   - ✅ **Auto-Format Detection**: Supports MD, TXT, PDF, DOCX, HTML automatically
+   - ✅ **Smart Token Management**: LLM-based summarization if content exceeds 10k tokens
+
+   The feature is enabled by default in `config.yaml`. See [docs/MAIN_DOCUMENT_GUIDE.md](docs/MAIN_DOCUMENT_GUIDE.md) for advanced configuration.
+
+7. **Pull Ollama model**:
+   ```bash
+   ollama pull llama3.2:3b
+   # Or other small models: phi3:mini, gemma2:2b
+   ```
+
+## 🚀 Usage
+
+### Local Development
+
+1. **Build the retrieval index** (default: hybrid BM25 + Vector):
+   ```bash
+   # Build with default strategy from config (bm25_vector)
+   python -m src.build_vectorstore
+
+   # Or specify a strategy explicitly
+   python -m src.build_vectorstore --strategy bm25_vector  # Hybrid (recommended)
+   python -m src.build_vectorstore --strategy vector       # Vector-only
+   python -m src.build_vectorstore --strategy bm25         # BM25-only
+   ```
+
+2. **Run the Streamlit app**:
+   ```bash
+   streamlit run app.py
+   ```
+
+3. **Open browser**: Navigate to `http://localhost:8501`
+
+**Screenshot Example:**
+
+<p align="center">
+  <img src="docs/screenshot-example.png" alt="ProfillyBot UI Example" width="800" />
+</p>
+<p align="center"><em>Streamlit chatbot interface in action</em></p>
+
+### 🐳 Docker Compose
+
+Run the complete stack with Docker Compose (includes Ollama):
+
+```bash
+# Copy environment template
+cp .env.template .env
+
+# Start all services (app + Ollama)
+docker compose up -d
+
+# View logs
+docker compose logs -f
+
+# Stop services
+docker compose down
+```
+
+Access the app at `http://localhost:7860`
+
+### 📋 Makefile Commands
+
+Use the Makefile for common development tasks:
+
+```bash
+make help          # Show all available commands
+```
+
+| Category | Command | Description |
+|----------|---------|-------------|
+| **Setup** | `make setup` | Initial project setup (copy env, install deps) |
+| | `make install` | Install dependencies |
+| | `make install-dev` | Install with dev dependencies |
+| **Development** | `make run` | Run Streamlit app locally |
+| | `make index-build` | Build vector and BM25 indexes |
+| | `make index-rebuild` | Rebuild indexes from scratch |
+| **Quality** | `make test` | Run tests |
+| | `make test-cov` | Run tests with coverage |
+| | `make lint` | Run linter (ruff) |
+| | `make format` | Format code |
+| | `make check` | Run all checks |
+| **Docker** | `make docker-up` | Start all services |
+| | `make docker-down` | Stop all services |
+| | `make docker-rebuild` | Rebuild and restart |
+| | `make docker-logs` | Show logs (follow mode) |
+| | `make docker-clean` | Stop and remove volumes |
+| **Ollama** | `make ollama-pull` | Pull configured model |
+| | `make ollama-list` | List available models |
+| **Cleanup** | `make clean` | Remove build artifacts |
+| | `make clean-all` | Clean everything + indexes |
+
+## 🌐 Deployment to Hugging Face Spaces (ZeroGPU)
+
+ProfillyBot on Spaces uses **Gradio + ZeroGPU** with `Qwen/Qwen2.5-3B-Instruct`
+(`gradio_app.py`). Streamlit (`app.py`) remains available for local/Ollama/Groq.
+
+### Live Space
+
+- [huggingface.co/spaces/MinhDS/ProfillyBot](https://huggingface.co/spaces/MinhDS/ProfillyBot)
+
+### Deploy / update
+
+1. Create a Gradio Space with **ZeroGPU** hardware (Pro account required to *create* ZeroGPU Spaces).
+2. Ensure this repo has:
+   - `data/documents/MY_CV_2_0.pdf` (or your profile docs)
+   - Pre-built `chroma_db/` + `bm25_index/` (recommended) — or let the app build on first start
+3. Push to the Space:
+   ```bash
+   git remote add hf https://huggingface.co/spaces/MinhDS/ProfillyBot
+   git push hf main
+   ```
+4. Confirm Space Settings: SDK **Gradio**, hardware **ZeroGPU**, app file `gradio_app.py`.
+
+### Important Notes for HF Spaces
+
+- **ZeroGPU is Gradio-only** — do not use the Docker/Ollama `Dockerfile` on ZeroGPU
+- **SLM**: `Qwen/Qwen2.5-3B-Instruct` via the `transformers` provider
+- **Embeddings**: `all-MiniLM-L6-v2` on CPU (does not consume GPU quota)
+- **Vector DB**: Pre-build indexes locally when possible for faster cold start
+- **Secrets**: optional `HF_TOKEN` only if you switch to a gated model
+
+## ⚙️ Configuration
+
+### `config.yaml` - Main Settings
+
+```yaml
+profile:
+  name: "Your Name"  # ← Change this!
+  title: "Your Title"
+
+llm:
+  model: "llama3.2:3b"  # Choose your model
+  temperature: 0.7
+
+# Retrieval Strategy Configuration
+retrieval:
+  strategy: "bm25_vector"  # Options: vector, bm25, bm25_vector
+  final_k: 4               # Number of documents to return
+
+  vector:
+    search_type: "similarity"
+    k: 10                  # Docs to retrieve before fusion
+
+  bm25:
+    k: 10                  # Docs to retrieve before fusion
+    tokenizer: "simple"
+
+  fusion:
+    algorithm: "rrf"       # Reciprocal Rank Fusion
+    weights:
+      vector: 0.7          # 70% weight on semantic search
+      bm25: 0.3            # 30% weight on keyword search
+
+document_processing:
+  chunk_size: 1000
+  chunk_overlap: 200
+```
+
+### `.env` - Environment Variables
+
+```bash
+OLLAMA_BASE_URL=http://localhost:11434
+CHROMA_PERSIST_DIR=./chroma_db
+```
+
+## 📚 Project Structure
+
+```
+profillybot/
+├── app.py                          # Streamlit app entry point
+├── pyproject.toml                  # UV/pip dependencies & config
+├── config.yaml                     # RAG & LLM settings
+├── .env.template                   # Environment variables template
+├── Makefile                        # Development commands
+├── docker-compose.yml              # Docker Compose configuration
+├── Dockerfile                      # Docker image (standalone with Ollama)
+├── Dockerfile.app                  # Docker image (for Compose)
+├── README.md
+├── data/
+│   └── documents/                  # Your profile documents
+├── src/
+│   ├── __init__.py
+│   ├── document_processor.py       # Load & chunk documents
+│   ├── vectorstore.py              # ChromaDB operations
+│   ├── llm_handler.py              # Ollama/LLM interface
+│   ├── rag_pipeline.py             # RAG chain logic
+│   ├── main_document_loader.py     # Main document management
+│   ├── response_enhancer.py        # Response post-processing
+│   ├── config_loader.py            # Load config.yaml & .env
+│   ├── build_vectorstore.py        # CLI to build indexes
+│   └── retrieval/                  # Extensible retrieval system
+│       ├── __init__.py
+│       ├── base.py                 # BaseRetrieverStrategy ABC
+│       ├── factory.py              # RetrieverFactory
+│       ├── fusion.py               # RRF and fusion algorithms
+│       ├── stores/
+│       │   └── bm25_store.py       # BM25 index storage
+│       └── strategies/
+│           ├── vector.py           # Vector-only strategy
+│           ├── bm25.py             # BM25-only strategy
+│           └── bm25_vector.py      # Hybrid BM25+Vector strategy
+├── chroma_db/                      # Vector database (auto-generated)
+├── bm25_index/                     # BM25 index (auto-generated)
+└── tests/                          # Unit tests
+```
+
+## 🎯 Recommended Models for HF Spaces
+
+| Model | Size | Speed | Quality | HF Spaces Tier |
+|-------|------|-------|---------|----------------|
+| `llama3.2:3b` | 3B | Fast | Good | Free ✅ |
+| `phi3:mini` | 3.8B | Fast | Good | Free ✅ |
+| `gemma2:2b` | 2B | Very Fast | Decent | Free ✅ |
+| `llama3.1:8b` | 8B | Medium | Great | Upgraded GPU |
+
+## 🔍 LLM Tracing with LangSmith
+
+Monitor prompts, responses, and latency using [LangSmith](https://smith.langchain.com).
+
+### Setup
+
+1. **Get API Key**: Sign up at [smith.langchain.com](https://smith.langchain.com) (free tier: 5,000 traces/month)
+
+2. **Configure `.env`**:
+   ```bash
+   LANGCHAIN_TRACING_V2=true
+   LANGCHAIN_API_KEY=your_api_key_here
+   LANGCHAIN_PROJECT=profillybot
+   ```
+
+3. **Restart the app**:
+   ```bash
+   streamlit run app.py
+   ```
+
+### What You Can See
+
+- Full prompts sent to LLM (system prompt + context + question)
+- Complete LLM responses
+- Latency breakdown per step
+- Token usage (when available)
+
+## 🔧 Troubleshooting
+
+### Ollama Connection Issues
+```bash
+# Check Ollama is running
+ollama list
+
+# Start Ollama service
+ollama serve
+```
+
+### ChromaDB / Index Errors
+```bash
+# Rebuild all indexes from scratch
+rm -rf chroma_db/ bm25_index/
+python -m src.build_vectorstore --force-rebuild
+```
+
+### HuggingFace Spaces Issues
+- Check logs in the Space's "Logs" tab
+- Ensure `requirements.txt` is generated: `uv pip compile pyproject.toml -o requirements.txt`
+- Verify GPU/CPU settings match your model size
+
+## 📄 License
+
+MIT License - see LICENSE file
+
+---
+
+**Note**: Remember to update `config.yaml` with your personal information before deploying!

SETUP_GUIDE.md

@@ -0,0 +1,371 @@
+# 🚀 Setup Guide - ProfillyBot
+
+This guide will walk you through setting up ProfillyBot, your personal profile chatbot, from scratch.
+
+## 📋 Prerequisites
+
+### 1. Install Python 3.10+
+
+```bash
+# Check your Python version
+python --version  # Should be 3.10 or higher
+```
+
+### 2. Install UV Package Manager
+
+```bash
+# Linux/macOS
+curl -LsSf https://astral.sh/uv/install.sh | sh
+
+# Windows (PowerShell)
+irm https://astral.sh/uv/install.ps1 | iex
+```
+
+### 3. Install Ollama
+
+Download and install from [ollama.ai](https://ollama.ai)
+
+```bash
+# Verify installation
+ollama --version
+```
+
+## 🛠️ Local Setup
+
+### Step 1: Clone and Install
+
+```bash
+# Clone the repository
+git clone https://github.com/Tin-Hoang/slm-profile-rag.git
+cd slm-profile-rag  # or rename the folder to profillybot if you prefer
+
+# Create virtual environment with UV
+uv venv
+
+# Activate virtual environment
+# Linux/macOS:
+source .venv/bin/activate
+# Windows:
+# .venv\Scripts\activate
+
+# Install dependencies
+uv pip install -r requirements.txt
+```
+
+### Step 2: Configuration
+
+#### 2.1 Environment Variables
+
+```bash
+# Copy environment template
+cp env.template .env
+
+# Edit .env file (optional, defaults work for local development)
+# Update OLLAMA_BASE_URL if needed
+```
+
+#### 2.2 Configure Your Profile
+
+Edit `config.yaml`:
+
+```yaml
+profile:
+  name: "Your Full Name"  # ⚠️ CHANGE THIS
+  title: "Your Professional Title"  # ⚠️ CHANGE THIS
+  greeting: "Hi! I'm ProfillyBot, trained on {name}'s professional background..."
+
+llm:
+  model: "llama3.2:3b"  # Choose based on your hardware
+  temperature: 0.7
+
+# Adjust other settings as needed
+```
+
+### Step 3: Pull LLM Model
+
+```bash
+# Pull the model specified in config.yaml
+ollama pull llama3.2:3b
+
+# Or try other models:
+# ollama pull phi3:mini       # Microsoft Phi-3 (3.8B)
+# ollama pull gemma2:2b       # Google Gemma 2 (2B)
+# ollama pull llama3.1:8b     # Larger, better quality (if you have GPU)
+```
+
+### Step 4: Add Your Documents
+
+```bash
+# Navigate to documents directory
+cd data/documents
+
+# Copy your profile documents here
+# Supported formats: PDF, DOCX, HTML, TXT, MD
+```
+
+**Recommended Documents:**
+- ✅ Resume/CV (PDF or DOCX)
+- ✅ LinkedIn profile (export as PDF)
+- ✅ Project reports and case studies
+- ✅ Portfolio descriptions
+- ✅ Publications, certifications
+- ✅ Cover letters, personal statements
+
+**Tips:**
+- Use descriptive filenames
+- Ensure documents are well-structured with headings
+- Include detailed information (the more context, the better!)
+- Remove the sample `SAMPLE_README.md` file
+
+### Step 5: Build Vector Store
+
+```bash
+# Make sure you're in the project root
+cd ../..  # if you're still in data/documents
+
+# Build the vector database
+python -m src.build_vectorstore
+
+# Output should show:
+# - Processing documents from: ./data/documents
+# - Successfully processed X document chunks
+# - ✅ Vector store built successfully!
+```
+
+**Troubleshooting:**
+- If no documents found: Check `data/documents/` directory
+- If import errors: Ensure virtual environment is activated
+- If memory issues: Reduce chunk_size in `config.yaml`
+
+### Step 6: Start Ollama (if not running)
+
+```bash
+# Start Ollama server
+ollama serve
+```
+
+Keep this terminal running, or run it as a background service.
+
+### Step 7: Run the Application
+
+```bash
+# In a new terminal (with virtual environment activated)
+streamlit run app.py
+
+# The app will open at: http://localhost:8501
+```
+
+## 🎨 Customization
+
+### Change UI Theme
+
+Edit `.streamlit/config.toml`:
+
+```toml
+[theme]
+primaryColor = "#FF4B4B"  # Change to your color
+backgroundColor = "#FFFFFF"
+```
+
+### Adjust RAG Settings
+
+Edit `config.yaml`:
+
+```yaml
+document_processing:
+  chunk_size: 1000      # Increase for more context per chunk
+  chunk_overlap: 200    # Increase for better continuity
+
+vectorstore:
+  search_kwargs:
+    k: 4                # Number of chunks to retrieve (increase for more context)
+
+llm:
+  temperature: 0.7      # Lower (0.3) for factual, higher (0.9) for creative
+  max_tokens: 512       # Maximum response length
+```
+
+### Add Example Questions
+
+Edit `config.yaml`:
+
+```yaml
+ui:
+  example_questions:
+    - "What is {name}'s background?"
+    - "What programming languages does {name} know?"
+    - "Tell me about {name}'s biggest project"
+    # Add your own questions
+```
+
+### Customize System Prompt
+
+Edit `config.yaml`:
+
+```yaml
+llm:
+  system_prompt: |
+    You are an AI assistant representing {name}.
+    [Customize your instructions here]
+```
+
+## 🌐 Deployment to Hugging Face Spaces
+
+### Option 1: Direct Upload
+
+1. Go to [Hugging Face Spaces](https://huggingface.co/spaces)
+2. Click "Create new Space"
+3. Choose "Streamlit" SDK
+4. Upload all project files
+5. Add your documents to `data/documents/`
+6. Space will automatically build and deploy
+
+### Option 2: Git Integration
+
+```bash
+# Initialize git (if not already)
+git init
+git add .
+git commit -m "Initial commit"
+
+# Add Hugging Face as remote
+git remote add hf https://huggingface.co/spaces/YOUR_USERNAME/SPACE_NAME
+
+# Push to deploy
+git push hf main
+```
+
+### Important for HF Spaces
+
+1. **Pre-build Vector Store**: Build locally and include `chroma_db/` in git
+
+```bash
+# Build locally
+python -m src.build_vectorstore
+
+# Remove chroma_db from .gitignore temporarily
+# Commit and push
+git add chroma_db/
+git commit -m "Add pre-built vector store"
+git push hf main
+```
+
+2. **Model Size**: Use small models (3B-7B params) for free tier
+   - Recommended: `llama3.2:3b`, `phi3:mini`, `gemma2:2b`
+
+3. **Persistent Storage**: Enable in Space settings for ChromaDB
+
+4. **Secrets**: Add API keys in Space Settings → Repository secrets
+
+## 🧪 Testing
+
+```bash
+# Install dev dependencies
+uv pip install pytest pytest-cov ruff
+
+# Run tests
+pytest
+
+# Run linter
+ruff check .
+
+# Format code
+ruff format .
+```
+
+## 🔧 Common Issues
+
+### Issue: "Ollama connection failed"
+
+**Solution:**
+```bash
+# Make sure Ollama is running
+ollama serve
+
+# Check if model is pulled
+ollama list
+
+# Pull model if missing
+ollama pull llama3.2:3b
+```
+
+### Issue: "Vector store not found"
+
+**Solution:**
+```bash
+# Rebuild vector store
+python -m src.build_vectorstore --force-rebuild
+```
+
+### Issue: "No documents found"
+
+**Solution:**
+- Check `data/documents/` has files
+- Verify file extensions are supported (.pdf, .docx, .html, .txt, .md)
+- Check file permissions
+
+### Issue: "Out of memory"
+
+**Solution:**
+```yaml
+# Reduce chunk size in config.yaml
+document_processing:
+  chunk_size: 500  # Reduced from 1000
+
+# Or use a smaller model
+llm:
+  model: "gemma2:2b"
+```
+
+### Issue: "Slow response times"
+
+**Solution:**
+- Use smaller model: `gemma2:2b` or `phi3:mini`
+- Reduce retrieval chunks: `k: 2` instead of `k: 4`
+- Lower temperature: `temperature: 0.5`
+- Enable GPU for Ollama (if available)
+
+## 📊 Performance Optimization
+
+### For Local Development
+
+```yaml
+# config.yaml
+embeddings:
+  device: "cuda"  # If you have GPU
+
+llm:
+  model: "llama3.1:8b"  # Better quality with GPU
+```
+
+### For Production/HF Spaces
+
+```yaml
+embeddings:
+  device: "cpu"
+  model_name: "sentence-transformers/all-MiniLM-L6-v2"  # Fast, lightweight
+
+llm:
+  model: "llama3.2:3b"  # Best balance for CPU
+  temperature: 0.6
+  max_tokens: 384  # Faster responses
+```
+
+## 🎯 Next Steps
+
+1. ✅ Customize your profile information
+2. ✅ Add comprehensive documents
+3. ✅ Test with various questions
+4. ✅ Deploy to Hugging Face Spaces
+5. ✅ Share the link with recruiters!
+
+## 📞 Support
+
+- Check [README.md](README.md) for general information
+- Review [config.yaml](config.yaml) for all options
+- See [tests/](tests/) for examples
+- Open an issue for bugs
+
+---
+
+**Happy chatting! 🚀**

app.py

@@ -0,0 +1,585 @@
+"""Streamlit app for ProfillyBot."""
+
+import logging
+import os
+import sys
+from datetime import datetime
+from pathlib import Path
+
+import streamlit as st
+
+# Add src to path
+sys.path.insert(0, str(Path(__file__).parent))
+
+from src.config_loader import get_config
+from src.llm_handler import (
+    detect_default_provider,
+    get_available_groq_models,
+    get_available_ollama_models,
+    get_default_groq_model,
+    get_default_ollama_model,
+    get_llm_handler,
+)
+from src.rag_pipeline import RAGPipeline
+
+# Configure logging
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+)
+logger = logging.getLogger(__name__)
+
+# Log LangSmith tracing status
+if os.getenv("LANGCHAIN_TRACING_V2", "").lower() == "true":
+    logger.info(
+        f"LangSmith tracing enabled for project: {os.getenv('LANGCHAIN_PROJECT', 'profillybot')}"
+    )
+
+
+def init_session_state():
+    """Initialize Streamlit session state."""
+    if "messages" not in st.session_state:
+        st.session_state.messages = []
+
+    if "rag_pipeline" not in st.session_state:
+        st.session_state.rag_pipeline = None
+
+    if "config" not in st.session_state:
+        st.session_state.config = None
+
+    if "pending_question" not in st.session_state:
+        st.session_state.pending_question = None
+
+    # LLM Provider settings
+    if "llm_provider" not in st.session_state:
+        st.session_state.llm_provider = detect_default_provider()
+
+    if "groq_api_key" not in st.session_state:
+        st.session_state.groq_api_key = ""
+
+    if "groq_model" not in st.session_state:
+        st.session_state.groq_model = get_default_groq_model()
+
+    if "ollama_model" not in st.session_state:
+        st.session_state.ollama_model = get_default_ollama_model()
+
+
+def load_config():
+    """Load configuration."""
+    if st.session_state.config is None:
+        try:
+            st.session_state.config = get_config()
+            logger.info("Configuration loaded successfully")
+        except Exception as e:
+            st.error(f"Error loading configuration: {e}")
+            st.stop()
+
+
+def load_rag_pipeline(force_reload: bool = False):
+    """Load RAG pipeline with current provider settings.
+
+    Args:
+        force_reload: Force reload even if pipeline exists
+    """
+    if st.session_state.rag_pipeline is None or force_reload:
+        with st.spinner("Loading RAG pipeline... This may take a moment."):
+            try:
+                # Get provider settings from session state
+                provider = st.session_state.llm_provider
+                api_key = st.session_state.groq_api_key or None
+                model = None
+
+                if provider == "groq":
+                    model = st.session_state.groq_model
+                elif provider == "ollama":
+                    model = st.session_state.ollama_model
+
+                # Create LLM handler with overrides
+                llm_handler = get_llm_handler(
+                    provider_override=provider,
+                    api_key_override=api_key,
+                    model_override=model,
+                )
+
+                st.session_state.rag_pipeline = RAGPipeline(llm_handler=llm_handler)
+                logger.info(f"RAG pipeline loaded with provider: {provider}")
+
+            except FileNotFoundError:
+                st.error(
+                    "❌ Vector store not found! Please build it first:\n\n"
+                    "```bash\n"
+                    "python -m src.build_vectorstore\n"
+                    "```"
+                )
+                st.info(
+                    "Make sure you have:\n"
+                    "1. Added your documents to `data/documents/`\n"
+                    "2. Run the build command above\n"
+                    "3. Started Ollama and pulled the model"
+                )
+                st.stop()
+
+            except Exception as e:
+                error_str = str(e).lower()
+
+                # Check if it's a Groq API error - try fallback to Ollama
+                if st.session_state.llm_provider == "groq" and (
+                    "401" in error_str
+                    or "invalid" in error_str
+                    or "api_key" in error_str
+                    or "429" in error_str
+                    or "rate" in error_str
+                ):
+                    logger.warning(f"Groq API error: {e}. Falling back to Ollama.")
+
+                    # Show warning to user
+                    st.warning(
+                        "⚠️ **Groq API Error** - Falling back to Ollama\n\n"
+                        "Possible causes:\n"
+                        "- Invalid API key\n"
+                        "- Rate limit exceeded (free tier cap)\n"
+                        "- API key expired\n\n"
+                        "**Get a new key at:** https://console.groq.com\n\n"
+                        "Using local Ollama instead..."
+                    )
+
+                    # Fallback to Ollama
+                    try:
+                        st.session_state.llm_provider = "ollama"
+                        ollama_model = st.session_state.ollama_model or get_default_ollama_model()
+
+                        llm_handler = get_llm_handler(
+                            provider_override="ollama",
+                            model_override=ollama_model,
+                        )
+
+                        st.session_state.rag_pipeline = RAGPipeline(llm_handler=llm_handler)
+                        logger.info("Fallback to Ollama successful")
+                        st.success(f"✅ Now using Ollama with model: {ollama_model}")
+                        return
+
+                    except Exception as fallback_error:
+                        st.error(f"Fallback to Ollama also failed: {fallback_error}")
+                        st.info(
+                            "Make sure Ollama is running:\n"
+                            "- Start Ollama: `ollama serve`\n"
+                            "- Pull a model: `ollama pull llama3.2:3b`"
+                        )
+                        logger.error(f"Ollama fallback error: {fallback_error}")
+                        st.stop()
+
+                # Not a Groq error or other error
+                st.error(f"Error loading RAG pipeline: {e}")
+                if st.session_state.llm_provider == "groq":
+                    st.info(
+                        "Groq issues:\n"
+                        "- Check your API key is valid\n"
+                        "- Get a free key at https://console.groq.com"
+                    )
+                else:
+                    st.info(
+                        "Ollama issues:\n"
+                        "- Ollama not running: `ollama serve`\n"
+                        "- Model not pulled: `ollama pull llama3.2:3b`"
+                    )
+                logger.error(f"RAG pipeline error: {e}", exc_info=True)
+                st.stop()
+
+
+def configure_page():
+    """Configure Streamlit page settings."""
+    config = st.session_state.config
+
+    page_title = config.get("ui.page_title", "💬 ProfillyBot")
+    page_icon = config.get("ui.page_icon", "🤖")
+    layout = config.get("ui.layout", "centered")
+
+    st.set_page_config(page_title=page_title, page_icon=page_icon, layout=layout)
+
+
+def display_header():
+    """Display app header."""
+    config = st.session_state.config
+
+    profile_name = config.get("profile.name", "Candidate")
+    profile_title = config.get("profile.title", "Professional")
+    greeting = config.format_template(config.get("profile.greeting", ""))
+
+    st.title(f"💬 ProfillyBot — Chat with {profile_name}'s Profile")
+    st.markdown(f"**{profile_title}**")
+
+    if greeting:
+        st.info(greeting)
+
+    st.divider()
+
+
+def display_sidebar():
+    """Display sidebar with information and controls."""
+    config = st.session_state.config
+
+    with st.sidebar:
+        st.header("ℹ️ About")
+
+        profile_name = config.get("profile.name", "the candidate")
+        st.markdown(
+            f"This chatbot can answer questions about **{profile_name}'s** "
+            "professional background, skills, experience, and projects."
+        )
+
+        st.divider()
+
+        # Example questions
+        st.subheader("💡 Example Questions")
+        example_questions = config.get("ui.example_questions", [])
+
+        for question in example_questions:
+            formatted_q = config.format_template(question)
+            if st.button(formatted_q, key=f"example_{hash(formatted_q)}"):
+                st.session_state.pending_question = formatted_q
+                st.rerun()
+
+        st.divider()
+
+        # LLM Provider Selection
+        st.subheader("🤖 LLM Provider")
+
+        # Provider selector
+        providers = ["ollama", "groq"]
+        current_provider_idx = (
+            providers.index(st.session_state.llm_provider)
+            if st.session_state.llm_provider in providers
+            else 0
+        )
+
+        new_provider = st.radio(
+            "Select Provider",
+            providers,
+            index=current_provider_idx,
+            format_func=lambda x: "🦙 Ollama (Local)" if x == "ollama" else "⚡ Groq (Cloud)",
+            key="provider_radio",
+            help="Ollama runs locally. Groq requires an API key but is faster.",
+        )
+
+        # Groq-specific settings
+        if new_provider == "groq":
+            st.markdown("---")
+            groq_key = st.text_input(
+                "Groq API Key",
+                value=st.session_state.groq_api_key,
+                type="password",
+                placeholder="Enter your Groq API key",
+                help="Get a free key at https://console.groq.com",
+                key="groq_key_input",
+            )
+
+            groq_models = get_available_groq_models()
+            current_model_idx = (
+                groq_models.index(st.session_state.groq_model)
+                if st.session_state.groq_model in groq_models
+                else 0
+            )
+
+            new_model = st.selectbox(
+                "Groq Model",
+                groq_models,
+                index=current_model_idx,
+                key="groq_model_select",
+            )
+
+            # Check if settings changed
+            settings_changed = (
+                new_provider != st.session_state.llm_provider
+                or groq_key != st.session_state.groq_api_key
+                or new_model != st.session_state.groq_model
+            )
+
+            if settings_changed and st.button("🔄 Apply Changes", type="primary"):
+                st.session_state.llm_provider = new_provider
+                st.session_state.groq_api_key = groq_key
+                st.session_state.groq_model = new_model
+                st.session_state.rag_pipeline = None  # Force reload
+                st.rerun()
+        else:
+            # Ollama selected - show model selector
+            st.markdown("---")
+            ollama_models = get_available_ollama_models()
+            current_ollama_idx = (
+                ollama_models.index(st.session_state.ollama_model)
+                if st.session_state.ollama_model in ollama_models
+                else 0
+            )
+
+            new_ollama_model = st.selectbox(
+                "Ollama Model",
+                ollama_models,
+                index=current_ollama_idx,
+                key="ollama_model_select",
+                help="Make sure the model is pulled: ollama pull <model>",
+            )
+
+            settings_changed = (
+                new_provider != st.session_state.llm_provider
+                or new_ollama_model != st.session_state.ollama_model
+            )
+
+            if settings_changed and st.button("🔄 Apply Changes", type="primary"):
+                st.session_state.llm_provider = new_provider
+                st.session_state.ollama_model = new_ollama_model
+                st.session_state.rag_pipeline = None
+                st.rerun()
+
+        # Show current active settings
+        st.markdown("---")
+        st.caption(f"**Active:** {st.session_state.llm_provider.upper()}")
+        if st.session_state.llm_provider == "groq":
+            st.caption(f"**Model:** {st.session_state.groq_model}")
+        else:
+            st.caption(f"**Model:** {st.session_state.ollama_model}")
+
+        st.divider()
+
+        # Clear chat button
+        if st.button("🗑️ Clear Chat History"):
+            st.session_state.messages = []
+            st.rerun()
+
+        # Settings
+        with st.expander("⚙️ Advanced Settings"):
+            st.markdown("**Vector Store**")
+            st.markdown(f"- Collection: {config.get('vectorstore.collection_name', 'N/A')}")
+            st.markdown(f"- Top K: {config.get('vectorstore.search_kwargs.k', 'N/A')}")
+
+            st.markdown("**Document Processing**")
+            st.markdown(f"- Chunk Size: {config.get('document_processing.chunk_size', 'N/A')}")
+            st.markdown(f"- Overlap: {config.get('document_processing.chunk_overlap', 'N/A')}")
+
+
+def extract_chat_history(messages: list, exclude_last: bool = True) -> list[dict]:
+    """Extract and format chat history for LLM.
+
+    Args:
+        messages: List of Streamlit message dicts
+        exclude_last: Whether to exclude the last message (usually the current one being processed)
+
+    Returns:
+        List of formatted message dicts with 'role' and 'content' keys
+    """
+    if not messages:
+        return []
+
+    # Exclude the last message if requested (current user message being processed)
+    history_messages = messages[:-1] if exclude_last else messages
+
+    # Filter to only user and assistant messages, format for LLM
+    formatted_history = []
+    for msg in history_messages:
+        role = msg.get("role")
+        content = msg.get("content", "")
+
+        if role in ("user", "assistant") and content:
+            formatted_history.append({"role": role, "content": content})
+
+    return formatted_history
+
+
+def truncate_chat_history(
+    history: list[dict], max_turns: int | None = None, max_tokens: int | None = None
+) -> list[dict]:
+    """Truncate chat history based on turn count or token limit.
+
+    Args:
+        history: List of message dicts
+        max_turns: Maximum number of Q&A pairs to keep (None = no limit)
+        max_tokens: Maximum tokens allowed (None = no limit)
+
+    Returns:
+        Truncated history list
+    """
+    from src.main_document_loader import get_main_document_loader
+
+    if not history:
+        return []
+
+    # Apply turn-based truncation first (keep most recent)
+    if max_turns is not None and max_turns > 0:
+        # Each turn = 1 user + 1 assistant message
+        max_messages = max_turns * 2
+        history = history[-max_messages:] if len(history) > max_messages else history
+
+    # Apply token-based truncation if needed
+    if max_tokens is not None and max_tokens > 0:
+        loader = get_main_document_loader()
+        current_tokens = 0
+        truncated_history = []
+
+        # Go through history in reverse, adding messages until we hit token limit
+        for msg in reversed(history):
+            content = msg.get("content", "")
+            msg_tokens = loader.count_tokens(content)
+
+            if current_tokens + msg_tokens <= max_tokens:
+                truncated_history.insert(0, msg)
+                current_tokens += msg_tokens
+            else:
+                break
+
+        return truncated_history
+
+    return history
+
+
+def display_chat_messages():
+    """Display chat messages."""
+    config = st.session_state.config
+    show_timestamps = config.get("ui.show_timestamps", True)
+
+    for message in st.session_state.messages:
+        with st.chat_message(message["role"]):
+            st.markdown(message["content"])
+
+            # Display timestamp if enabled
+            if show_timestamps and "timestamp" in message:
+                st.caption(f"_{message['timestamp']}_")
+
+            # Display sources if available
+            if "sources" in message and message["sources"]:
+                with st.expander("📚 Sources"):
+                    st.markdown(message["sources"])
+
+
+def process_user_input(prompt: str):
+    """Process user input and generate response.
+
+    Args:
+        prompt: User's question
+    """
+    config = st.session_state.config
+    rag_pipeline = st.session_state.rag_pipeline
+
+    # Add user message
+    timestamp = datetime.now().strftime("%H:%M:%S")
+    st.session_state.messages.append({"role": "user", "content": prompt, "timestamp": timestamp})
+
+    # Display user message
+    with st.chat_message("user"):
+        st.markdown(prompt)
+        if config.get("ui.show_timestamps", True):
+            st.caption(f"_{timestamp}_")
+
+    # Extract and prepare chat history
+    chat_history = []
+    if config.get("chat.enable_history", True):
+        # Extract history excluding the current message we just added
+        raw_history = extract_chat_history(st.session_state.messages, exclude_last=True)
+
+        # Truncate based on configuration
+        max_turns = config.get("chat.max_history_turns", 10)
+        max_tokens = config.get("chat.max_history_tokens", 2000)
+
+        chat_history = truncate_chat_history(
+            raw_history,
+            max_turns=max_turns if max_turns > 0 else None,
+            max_tokens=max_tokens if max_tokens > 0 else None,
+        )
+
+        if chat_history:
+            logger.debug(f"Including {len(chat_history)} messages from chat history")
+
+    # Generate streaming response
+    with st.chat_message("assistant"):
+        try:
+            # Create placeholder for streaming content
+            response_placeholder = st.empty()
+
+            # Stream the response token by token with spinner
+            streamed_content = ""
+            with st.spinner("Thinking (this may take a few minutes on a free-tier HF space)..."):
+                for chunk in rag_pipeline.stream_answer(prompt, chat_history=chat_history):
+                    streamed_content += chunk
+                    response_placeholder.markdown(streamed_content + "▌")
+
+            # Remove cursor and show final streamed content
+            response_placeholder.markdown(streamed_content)
+
+            # Enhance the response for better tone (post-stream enhancement)
+            answer = streamed_content
+            if config.get("rag.enhance_responses", True):
+                from src.response_enhancer import get_response_enhancer
+
+                enhancer = get_response_enhancer()
+                enhanced_answer = enhancer.enhance_with_context(streamed_content, prompt)
+                if enhanced_answer != streamed_content:
+                    answer = enhanced_answer
+                    # Update display with enhanced version
+                    response_placeholder.markdown(answer)
+                    logger.debug("Response enhanced for better tone")
+
+            # Get source documents (after streaming completes)
+            sources = rag_pipeline.get_source_documents(prompt, chat_history=chat_history)
+            sources_text = ""
+            if config.get("rag.include_sources", True) and sources:
+                sources_text = rag_pipeline.format_sources(sources)
+                with st.expander("📚 Sources"):
+                    st.markdown(sources_text)
+
+            # Add to message history
+            timestamp = datetime.now().strftime("%H:%M:%S")
+            st.session_state.messages.append(
+                {
+                    "role": "assistant",
+                    "content": answer,
+                    "timestamp": timestamp,
+                    "sources": sources_text,
+                }
+            )
+
+        except Exception as e:
+            error_msg = f"I apologize, but I encountered an error: {e}"
+            st.error(error_msg)
+            logger.error(f"Error generating response: {e}", exc_info=True)
+
+            st.session_state.messages.append(
+                {"role": "assistant", "content": error_msg, "timestamp": timestamp}
+            )
+
+
+def main():
+    """Main application entry point."""
+    # Initialize session state
+    init_session_state()
+
+    # Load configuration
+    load_config()
+
+    # Configure page
+    configure_page()
+
+    # Load RAG pipeline
+    load_rag_pipeline()
+
+    # Display UI
+    display_header()
+    display_sidebar()
+
+    # Display chat messages
+    display_chat_messages()
+
+    # Process pending question from example button click
+    if st.session_state.pending_question:
+        pending = st.session_state.pending_question
+        st.session_state.pending_question = None
+        process_user_input(pending)
+        st.rerun()
+
+    # Chat input
+    if prompt := st.chat_input("Ask me anything about the profile..."):
+        process_user_input(prompt)
+
+    # Limit chat history
+    max_history = st.session_state.config.get("ui.max_chat_history", 50)
+    if len(st.session_state.messages) > max_history:
+        st.session_state.messages = st.session_state.messages[-max_history:]
+
+
+if __name__ == "__main__":
+    main()

bm25_index/bm25_index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ae5802b1436443f1b185dc4c4c1b25407932fa7a3a01429f52e766794176495
+size 25337

bm25_index/documents.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc6e6df6c9c7d9403681d373ad2944e7cf2d774268aba63884a0c5a8cac33a34
+size 14995

bm25_index/metadata.json

@@ -0,0 +1,5 @@
+{
+  "doc_hash": "cd60d29f92f5390d",
+  "num_documents": 10,
+  "tokenizer": "simple"
+}

bm25_index/tokenized_corpus.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4219624d37be72d6ff179b599692006d53bc43f3d4dd03a5b7861ff0e4f567
+size 16329

chroma_db/ae1acac2-515c-4deb-b9d0-675bd72bab8b/data_level0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d3b07c5f2f6b3b7e52a35ab5f91bef157094ca463032216b041d6f20e31699c
+size 167600

chroma_db/ae1acac2-515c-4deb-b9d0-675bd72bab8b/header.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0e81c3b22454233bc12d0762f06dcca48261a75231cf87c79b75e69a6c00150
+size 100

chroma_db/ae1acac2-515c-4deb-b9d0-675bd72bab8b/length.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ec18261d55fddaecb8c3d38325422e9e8c189edcc7508d0bedb6375ccc174b4
+size 400