// the find
GregaVrbancic/fastapi-celery
Minimal example utilizing fastapi and celery with RabbitMQ for task queue, Redis for celery backend and flower for monitoring the celery tasks.
A minimal reference project wiring FastAPI to Celery with RabbitMQ as the broker, Redis as the result backend, and Flower for task monitoring — all spun up via docker-compose. It exists purely as a copy-paste starting point, not a library or framework. Useful if you've never connected these four pieces before and want to see them talking to each other.
Whole stack runs with one command — no local installs of RabbitMQ or Redis needed. Includes a separate docker-compose-services.yml so you can run just the infrastructure and iterate on app code without Docker. poetry for dependency management is the right call over bare pip. Flower is pre-wired and exposed, which saves the annoying step of figuring out credentials and port-forwarding later.
The actual task code is a stub — there's nothing here about error handling, retries, task routing, or result expiry, which are the parts that actually bite you in production. No tests at all. Hypercorn as the ASGI server is an odd choice (uvicorn is the standard pairing with FastAPI) and no explanation is given. Last meaningful activity appears to be years old, so dependency versions are likely stale — you'll need to audit before using any of this in a real project.