Skip to content

Learn FastAPI from Django

A personal, opinionated set of conspect notes for a Django developer moving to FastAPI. Every topic explains FastAPI on its own terms and then puts it side by side with Django, so years of Django experience transfer instead of getting in the way.

All material targets FastAPI 0.139, Django 6.0, and Python 3.13 (with Pydantic v2, Starlette, and Uvicorn as the FastAPI stack).

Who this is for

You know Django well — models and migrations, the ORM and QuerySet, the admin, DRF serializers, class-based views, middleware, settings.py. You do not know FastAPI yet. This book assumes exactly that: it never explains what a migration is, but it explains every FastAPI, Starlette, and Pydantic concept from zero, and each article ends by mapping it back to what you already do in Django.

How the notes are organised

Each topic is a numbered folder. Inside each folder, conspects and runnable examples share one numeric sequence so the reading order is always obvious. The Django comparison lives in > **From Django:** callouts and side-by-side code, not in a separate chapter.

Getting started

  • What is FastAPI — the framework, the stack, and how its philosophy differs from Django's.

Planned topics

The structure below is the working outline for the book — it will grow into linked articles as each one is written.

  1. Getting started — what FastAPI is, installing it, your first app, the dev server (uvicorn vs runserver), project layout without Django's app/project scaffolding, and application lifespan (startup/shutdown, the AppConfig.ready() analogue).
  2. Routing and requests — path operations vs urls.py, path and query parameters, request bodies, responses and status codes, error handling and exceptions (HTTPException, handlers, the 422 shape), forms and file uploads, and the shift away from class-based views.
  3. Data with Pydantic — Pydantic models vs Django forms and DRF serializers, validation, response_model, and settings (pydantic-settings vs settings.py).
  4. Templates and static files — Jinja2Templates vs the Django Template Language, render() → returning an HTMLResponse, and serving static assets vs collectstatic.
  5. Databases — SQLAlchemy + Alembic: models, the session, migrations vs Django migrations, querying, serializing ORM objects with Pydantic (from_attributes, the ModelSerializer analogue), and async access.
  6. Dependency injection — Depends(), the idea Django has no direct equivalent for; dependencies with yield, sub-dependencies, reuse, and what replaces signals.
  7. Authentication — Django auth/sessions vs OAuth2/JWT, security dependencies, CSRF and the cookie-vs-token model, and CORS.
  8. Middleware — Starlette middleware vs Django's MIDDLEWARE stack.
  9. Async and concurrency — WSGI vs ASGI, async def and when to use it, background tasks vs Celery, and WebSockets vs Django Channels.
  10. Testing — pytest: TestClient, fixtures, and testing dependencies and async code.
  11. The admin gap and deployment — what replaces django.contrib.admin, and deploying Uvicorn workers.

Appendix — Ecosystem map: a Django-package → FastAPI-stack cheat-sheet (DRF → FastAPI-native, django-allauth → fastapi-users, Celery → arq/Celery, django.contrib.admin → SQLAdmin, django-filter → query params, caching, pagination, rate limiting, and packaging).

Companion projects

Every chapter ships with two small, runnable companion projects — one in FastAPI and one in Django — that implement exactly the functionality the chapter describes, with tests proving they work. You can read the chapter, then download and run both sides to see the same feature built each way.

Source