Report Generation Platform Backend Service

Welcome to the backend service of our Report Generation Platform! This powerful service automates research and report generation by searching internal documents, the web, user-uploaded files, and specified URLs, producing a comprehensive report with accurate citations.

🌟 Key Features

📈 Report Generation

Effortlessly generate reports by harnessing data from multiple sources:

User Input: Define your target audience, report objectives, and other relevant details.
Automated Research: The system conducts four types of research in parallel:
- Internal Documents Search: Leverage our database via Azure AI Search.
- Web Search: Utilize Tavily Search Engine for the most relevant online data.
- User-Specified URLs: Fetch data from URLs provided by you using Exa API.
- User-Uploaded Files: Perform semantic search on your uploaded documents using FAISS.
Ranking: Rank all retrieved chunks using a Large Language Model (LLM).
Selection and Generation: Pass the top 20 chunks to LLM for report generation, with optional manual selection.
Editing: Manually edit reports using a Tiptap editor.

🧠 AI Assistant

Our AI assistant is equipped to answer your queries using several advanced tools, and can even create charts based on your requirements:

Tavily Search Engine: For the latest web data.
Azure AI Search: To search internal document databases.
Exa API: For extracting information from user-specified URLs.
HighChart: For creating visualizations.

You can explore our demo video here: AI Assistant Demo

🛠 Technologies

FastAPI: Robust RESTful API framework
PostgreSQL: Reliable and powerful database
Alembic: Easy database migration
Azure OpenAI: Advanced large language model
Azure AI Search: Comprehensive vector database
Tavily Search Engine: State-of-the-art search engine
Langfuse: For LLM tracing and prompt management
Docker & Docker Compose: Simplified container management

🚀 Getting Started

Follow these steps to run the server locally:

Clone the Repository:

git clone https://github.com/VerdantVibes/fastapi-postgre.git

Build Docker Container:
```
make runBuildLocalDocer
```
Navigate to Backend Directory:
```
cd backend
```
Set Environment Variables:
- Copy env.example and rename it to .env.local
- Replace the placeholder values with your actual credentials.
Run the Server:
```
make runLocal
```
The server will be up at http://localhost:8001, and you can test the endpoints at http://localhost:8000/docs.

Demo Video

Enjoy using the Report Generation Platform! 🚀


Feel free to reach out if you have any questions or need further assistance. Happy developing! 😊