Update README.md

README.md

@@ -1,14 +1,94 @@
+# 🔬 AI Research Companion (Groq + LangChain + FAISS)
+
+An advanced **AI-powered research assistant** that helps you analyze academic papers, ask natural language questions, generate engaging summaries, and discover related research papers — all from a modern, tabbed **Gradio interface**.
+
+---
+
+## 🚀 Features
+
+✅ **PDF Upload & Text Extraction** – Extracts full text from research papers  
+✅ **Chunking & Vector Embedding** – Uses LangChain + HuggingFace embeddings for semantic search  
+✅ **Groq LLM Q&A** – Powered by `llama-3.3-70b-versatile` for accurate, context-aware answers  
+✅ **Cited Source References** – Displays the exact chunks used for each answer  
+✅ **Research Paper Summarization** – Creates engaging, layperson-friendly summaries  
+✅ **Similar Paper Discovery** – Queries arXiv API to find related academic works  
+✅ **Beautiful Multi-Tab UI** – Fully custom styled with Gradio + CSS  
+
 ---
-title: My-App
-emoji: 🚀
-colorFrom: blue
-colorTo: green
-sdk: gradio
-sdk_version: "4.44.0"
-app_file: app.py
-pinned: false
+
+## 🛠 Tech Stack
+
+- **Python 3.9+**
+- [Gradio](https://gradio.app/) – Interactive UI framework
+- [LangChain](https://www.langchain.com/) – Document processing & QA chain
+- [FAISS](https://github.com/facebookresearch/faiss) – Efficient similarity search
+- [HuggingFace Sentence Transformers](https://www.sbert.net/) – Embeddings (`all-mpnet-base-v2`)
+- [Groq API](https://groq.com/) – High-performance LLM inference
+- [PyPDF2](https://pypi.org/project/PyPDF2/) – PDF parsing
+- [Feedparser](https://pypi.org/project/feedparser/) – arXiv paper search
+- **Custom CSS** – Modern tabbed layout, shadows, gradients, and animations
+
+---
+
+## 📦 Installation & Setup
+
+### 1️⃣ Clone the Repository
+
+```bash
+git clone https://github.com/bobbythomas985/Research_Assistant
+cd Research_Assistant
+```
+### 2️⃣ Install Dependencies
+```bash
+pip install -r requirements.txt
+```
+### 3️⃣ Set Up Your API Key
+Export your Groq API key as an environment variable:
+**Linux / macOS**
+```bash
+export GROQ_API_KEY="your_api_key_here"
+```
+**Windows**
+```powershell
+setx GROQ_API_KEY "your_api_key_here"
+```
+Alternatively, replace the placeholder in **app.py**:
+```python
+GROQ_API_KEY = os.getenv("GROQ_API_KEY", "your-api-key")
+```
+### 4️⃣ Run the App
+```bash
+python app.py
+```
+
 ---
 
-# My Hugging Face Space
+## 🖥️ How It Works
+
+1️⃣ **Upload a PDF**  
+📄 The system extracts all text from the research paper.
+
+2️⃣ **Process & Embed**  
+🔍 Splits the extracted text into overlapping chunks and creates a **FAISS vector index** using **HuggingFace embeddings** for efficient semantic search.
+
+3️⃣ **Ask Questions**  
+❓ User questions are converted into embeddings and matched with the most relevant chunks from the document.
+
+4️⃣ **LLM Answer Generation**  
+🤖 Groq’s `llama-3.3-70b-versatile` model is used to generate accurate, context-aware answers with a custom prompt.
+
+5️⃣ **Summarize & Discover Papers**  
+📝 Generates engaging, structured summaries of the document and retrieves similar papers from **arXiv** for further reading.
+
+
+## 🔮 Future Improvements
+
+- 📚 **Multi-document support** – Build a single knowledge base from multiple PDFs  
+- 📷 **LLM Reranking** – Use cross-encoder reranking for better context selection  
+- 📑 **Clickable Source References** – Jump directly to relevant sections inside the PDF  
+- 🚀 **Deploy on Hugging Face Spaces / Streamlit Cloud** – Make it public and shareable  
+- 🌍 **Multilingual Q&A** – Integrate translation for global research accessibility
+
+---
 
-This is a demo Gradio application powered by Groq, LangChain, and FAISS.
+> *Empowering researchers to go from papers → insights → new discoveries.*