EduFlow RAG System

AI-powered retrieval and content generation platform for micro-learning workflows.

What EduFlow Does

• Ingests and indexes knowledgebase documents
• Retrieves relevant context for generation tasks
• Generates learning content in multiple formats (for example PDF and voice assets)
• Supports chatbot streaming, background generation jobs, and prompt versioning

Tech Stack

• Backend: FastAPI
• AI/Orchestration: OpenAI + Anthropic integrations, LangChain
• Storage/Search: Azure Blob Storage + Azure AI Search
• Data layer: Azure SQL (background jobs + prompt management)
• Frontend: dummy-ui (served separately)

Project Structure

eduflow/
├── app/                    # FastAPI application core
│   ├── adapters/           # External service adapters (Azure, model clients, etc.)
│   ├── core/               # Shared application logic
│   ├── knowledgebase/      # Indexing and document processing logic
│   ├── pipelines/          # Generation and chat workflow pipelines
│   ├── prompts/            # Prompt models and management services
│   ├── retrievers/         # Retrieval helpers
│   ├── routers/            # API route definitions
│   └── main.py             # API entrypoint
├── config/                 # Configuration and logging setup
├── docs/                   # Architecture diagrams and documentation
├── dummy-ui/               # Frontend app
├── evaluator_service/      # Evaluator microservice
├── tests/                  # Test suite
├── docker-compose.yml
├── requirements.txt
└── README.md

Quick Start

Prerequisites

• Python 3.10+ (recommended)
• Docker Desktop (optional, for containerized run)
• Azure resources configured for Blob/Search/SQL
• A populated .env file in the project root

Option 1: Run with Docker Compose

docker compose up --build

• Backend API: http://localhost:8000
• Frontend UI: http://localhost:3000

Option 2: Run backend locally

python -m venv .venv

# Windows PowerShell
.venv\Scripts\Activate.ps1

pip install -r requirements.txt
uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload

API Overview

Generation and Chat

• POST /chat/stream - stream chatbot responses using a selected index
• POST /gen/start - start background content generation
• POST /gen/start-baseline - start baseline generation flow
• POST /gen/start-memory - start memory-based generation flow
• POST /gen/start-test - trigger test generation flow

Knowledgebase

• POST /run-ai-indexing/{index_id} - provision indexing pipeline artifacts
• POST /knowledgebase/{index_id}/upload - upload and index documents
• DELETE /knowledgebase/{index_id}/documents - delete indexed docs by filename
• DELETE /index/{index_id} - remove index artifacts and related container
• GET /blobs - list blobs by container/prefix
• GET /test-blob-connection - test Blob connectivity

Background Jobs and Prompt Management

• GET /bg_jobs and GET /bg_jobs/{job_id} - inspect generation job state/results
• DELETE /bg_jobs/clear - clear background jobs
• POST /init-tables - initialize prompt tables
• GET /prompts/*, POST /prompts/new, POST /prompts/activate, PUT/DELETE /prompts/... - prompt lifecycle APIs

Development Notes

• CORS is currently open (allow_origins=["*"]) in app/main.py for development flexibility.
• requirements.txt includes weasyprint; on Windows, if install issues occur, reinstall cffi and fonttools as noted in the file.