backend

Backend Documentation

Backend-specific technical documentation for ComponentForge.

Core Documentation

• Backend Architecture - Complete backend system architecture, module organization, and patterns
• AI Pipeline - LangChain/LangGraph workflows, multi-agent orchestration, and LLM integration
• Database Schema - PostgreSQL schema, models, queries, and migration strategies

Additional Resources

• Generation Service - Code generation system architecture
• Prompting Guide - Prompt engineering best practices
• Monitoring - Observability and LangSmith setup
• Troubleshooting - Common issues and solutions

Technical Analysis

• Caching Analysis - Comprehensive analysis of caching strategies (prompt, embedding, result caching)
• Guardrails Analysis - Security guardrails assessment and implementation review

Quick Reference

Running the Backend

# Start Docker services
docker-compose up -d

# Activate virtual environment
cd backend && source venv/bin/activate

# Run development server
uvicorn src.main:app --reload

# Run tests
pytest tests/ -v

# Run migrations
alembic upgrade head

Key Technologies

• Framework: FastAPI with async/await
• AI: LangChain, LangGraph, LangSmith
• Database: PostgreSQL with SQLAlchemy (async)
• Vector DB: Qdrant for semantic search
• Cache: Redis for sessions and rate limiting
• Models: OpenAI GPT-4, GPT-4V, text-embedding-3-small

Architecture

The backend follows a layered architecture:

1. API Layer (src/api/v1/) - REST endpoints
2. Service Layer (src/services/) - Business logic
3. Agent Layer (src/agents/) - LangGraph AI agents
4. Data Layer (src/models/) - Database models
5. RAG Layer (src/rag/) - Vector operations

See Also

• Main Documentation
• API Reference
• Architecture Overview
• Features Documentation
• Deployment Guide
• Development Workflow