AI-Knowledge-Assistant

An intelligent chatbot using LangChain and OpenAI API, enabling document-based question answering.

Problem Statement

In organizations, valuable knowledge is often scattered across documents, reports, and PDFs—making it difficult for teams to retrieve relevant information quickly. Traditional search methods rely on keyword matching, which fails to capture semantic meaning and context.

The AI Knowledge Assistant solves this problem by combining Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs) to enable natural language question answering over custom documents. Users can upload PDFs or text files and interact with an intelligent chatbot that provides accurate, context-aware answers.

Tech Stack

Layer	Tools / Frameworks
Frontend	Streamlit (interactive chat UI)
Backend API	FastAPI (RESTful endpoints + Swagger docs)
AI / NLP	LangChain, OpenAI GPT API
Vector Database	FAISS (semantic embedding search)
Deployment	Docker, AWS EC2
Version Control	Git + GitHub Actions (CI/CD)

Features

📂 Document Upload: Upload PDFs or text files for ingestion.

🔍 Semantic Retrieval: Retrieve relevant document chunks using FAISS vector embeddings.

💬 Contextual Chat: Ask natural-language questions and receive answers grounded in your uploaded documents.

🧠 RAG Pipeline: Integrates retrieval with GPT-based generation for contextually accurate responses.

🧾 API Access: FastAPI backend exposes /upload and /query endpoints with Swagger documentation.

🐳 Dockerized Deployment: Fully containerized for easy deployment on any environment.

⚡ CI/CD Integration: Automated testing and deployment pipeline via GitHub Actions.

☁️ Scalable Hosting: Deployed on AWS EC2 instance with persistent storage.

System Architecture

            ┌────────────────────────────┐
            │         User Query          │
            └──────────────┬──────────────┘
                           │
                  ┌────────▼────────┐
                  │    Streamlit     │
                  │  (Chat UI App)   │
                  └────────┬────────┘
                           │ REST API Call
                  ┌────────▼────────┐
                  │     FastAPI      │
                  │ (Backend Server) │
                  └────────┬────────┘
                           │
        ┌──────────────────▼──────────────────┐
        │     LangChain RAG Pipeline          │
        │  (Embedding + Retrieval + LLM)      │
        └──────────────────┬──────────────────┘
                           │
                  ┌────────▼────────┐
                  │     FAISS DB     │
                  │ (Vector Store)   │
                  └──────────────────┘

Demo Screenshots

(To be add later)

Interface	Description
Chat UI	Upload documents and chat interactively
Swagger Docs	API documentation for upload and query endpoints
System Diagram	RAG workflow visualization

Setup and Installation

Clone the repository

git clone https://github.com//AI-Knowledge-Assistant.git cd AI-Knowledge-Assistant

Create a virtual environment

python -m venv venv source venv/bin/activate # On Mac/Linux venv\Scripts\activate # On Windows

Install dependencies

pip install -r requirements.txt

Run FastAPI backend

uvicorn app.backend.main:app --reload

Run Streamlit frontend

streamlit run app/frontend.py

Docker Deployment

Build and run using Docker Compose

docker-compose up --build

Access the app:

Streamlit UI → http://localhost:8501

FastAPI Docs → http://localhost:8000/docs

Future Enhancements

🔐 Add authentication (JWT-based access).

📚 Support multiple file types (CSV, DOCX, HTML).

🌍 Integrate Hugging Face local models for offline inference.

🧩 Add analytics dashboard (chat history, query performance).

🤝 Contributing

Contributions are welcome! Please fork the repo and submit a pull request with detailed changes.