Production-oriented hybrid recommendation engine built with FastAPI, vector retrieval, collaborative filtering, PostgreSQL, Redis, and ML adapter hooks for TensorFlow and PyTorch.
- FastAPI API surface for auth, recommendations, interactions, analytics, and monitoring.
- Hybrid recommendation logic using embeddings, vector similarity search, collaborative filtering, popularity, and engagement signals.
- FAISS-backed vector search path with deterministic NumPy fallback for local tests.
- PostgreSQL SQLAlchemy models and repository layer.
- Redis recommendation cache adapter and Celery worker wiring.
- Docker Compose stack with backend, worker, PostgreSQL, Redis, Prometheus, and Grafana.
- Benchmark scripts and resume-claim validation docs.
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
copy .env.example .env
uvicorn app.main:app --reloadAPI docs:
http://localhost:8000/docs
Demo credentials:
demo@example.com / ChangeMe123
analyst@example.com / ChangeMe123
copy .env.example .env
docker-compose up --buildServices:
- FastAPI:
http://localhost:8000 - Swagger:
http://localhost:8000/docs - Prometheus:
http://localhost:9090 - Grafana:
http://localhost:3000
python scripts/benchmark_relevance.py --k 3
python scripts/benchmark_latency.py --iterations 100 --top-k 10