rag-app

Example: RAG Application

This example demonstrates how to use AI Dev Team to build a Retrieval-Augmented Generation (RAG) application from scratch — fully autonomously.

What Gets Built

A production-ready RAG system with:

• FastAPI backend — document upload, chunking, embedding, semantic search, chat
• Angular 18 frontend — document management UI, chat interface
• MongoDB — document metadata, chat history
• ChromaDB — vector store for embeddings
• Claude API — answer generation with source citations
• Docker Compose — full containerized setup
• CI/CD — GitHub Actions pipeline with linting, tests, and Docker build

Quick Start

# 1. Initialize the project
python3 scripts/init_project.py "RAG App" \
  "Production-ready RAG system. Upload PDF/TXT/MD documents, chunk and embed them, \
   store in ChromaDB, query via natural language with Claude API. \
   FastAPI backend, Angular 18 frontend, MongoDB, Docker Compose."

# 2. (Optional) Write a detailed brief to the PO
cp examples/rag-app/po-brief.md board/inbox/po.md

# 3. (Optional) Set up GitLab/GitHub integration
cp config/.env.example config/.env
# Edit config/.env with your tokens
python3 scripts/init_integrations.py

# 4. Run the team
./scripts/orchestrator_cli.sh dev-team

# 5. Set up automated runs (optional)
crontab ai-team.crontab

What Happens

After the first cycle (~15-30 minutes):

• PO creates 8-10 user stories with acceptance criteria
• PM assigns stories to developers
• Dev1/Dev2 start implementing the backend scaffolding
• DevOps creates Dockerfile and docker-compose.yml

After 5-6 sprints (~50 cycles, ~1 night with 40-min cron):

• Full backend with 18+ services, 8 routers, middleware
• Angular frontend with chat, upload, collections UI
• 400+ tests with 80%+ coverage
• Docker Compose setup with health checks
• CI/CD pipeline
• Security audit completed

The PO Brief

See po-brief.md for the initial brief that was sent to the Product Owner agent. This is the only human input — everything else is autonomous.

Results

After running overnight, the team produced:

• 28 user stories across 6 sprints
• 26 completed, 2 in backlog
• ~3,600 lines of backend code
• 407+ tests across 39 test files
• 10 merged MRs on GitLab
• 0 agent failures

All managed through the GitLab board with issues, labels, and merge requests.