Python Celery FastAPI AI Integration

I am expanding an existing, research-heavy AI application and need a seasoned Python developer who can own the API layer of the platform. Your primary mission will be to design and implement a FastAPI-based RESTful interface that links our internal machine-learning pipelines to several external services, making sure the data flows securely, quickly, and reliably.

You will work side-by-side with data scientists and algorithm engineers; while they refine models, you will expose their work through well-structured endpoints and schedule long-running jobs with Celery. A deep understanding of asynchronous programming, background job queues, and the nuances of production-grade REST APIs is therefore essential.

Key tech you should already be fluent with includes:
• Python 3.x, FastAPI, Pydantic
• Celery with Redis or RabbitMQ back-ends
• OAuth2 / JWT for securing endpoints
• Docker and Git for reproducible builds and clean collaboration

Deliverables I expect:
• A versioned RESTful API that handles external service integration end-to-end
• Celery task queues for model execution and heavy I/O workloads
• Unit & integration test coverage ≥90 % plus API documentation (OpenAPI / Swagger)
• A concise deployment guide (Docker-compose or Helm chart)

Acceptance criteria:
1. All endpoints return <200 ms on average under load test of 500 rps.
2. Celery workers recover gracefully from network interruptions and retry tasks idempotently.
3. External service calls are abstracted behind clean interfaces to allow future swapping without breaking clients.
4. CI pipeline passes linting, tests, and type-checking on every merge request.

If architecting a fast, reliable bridge between AI models and the outside world sounds like your wheelhouse, let’s get started—the rest of the team is ready for your code.