Day 1: FastAPI Fundamentals: Build Your First API

Why FastAPI for AI Apps

When you build an AI application, you need a way for other programs to talk to it. That's what an API does — it's a defined interface for making requests and getting responses. FastAPI is the Python framework that makes building these interfaces fast, reliable, and automatic.

Why FastAPI specifically:

• Speed: One of the fastest Python web frameworks, comparable to Node.js and Go
• Automatic documentation: Visit /docs on any FastAPI server and you get a live interactive API reference — free, automatic, always up to date
• Type safety: Uses Python type hints to validate request and response data automatically
• Async support: Built for handling many requests concurrently — important when each request calls an AI API

Install and First Server

Install FastAPI and uvicorn (the server that runs it):

$ pip install fastapi uvicorn anthropic python-dotenv

Create main.py:

from fastapi import FastAPI

app = FastAPI()

@app.get("/")
def health_check():
    return {"status": "ok", "service": "AI API"}

Run it:

$ uvicorn main:app --reload

Open http://localhost:8000 — you get your first response. Open http://localhost:8000/docs — you get auto-generated documentation you can run right in the browser.

Add an AI Endpoint

Now build a real endpoint that accepts text and returns an AI-generated summary:

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
import anthropic

app = FastAPI(title="AI Text API")
client = anthropic.Anthropic()

class SummarizeRequest(BaseModel):
    text: str
    max_words: int = 100

@app.get("/")
def health_check():
    return {"status": "ok"}

@app.post("/summarize")
def summarize(req: SummarizeRequest):
    if not req.text.strip():
        raise HTTPException(status_code=400, detail="text cannot be empty")

    msg = client.messages.create(
        model="claude-sonnet-4-20250514",
        max_tokens=512,
        system=f"Summarize the text in {req.max_words} words or fewer. Return only the summary.",
        messages=[{"role": "user", "content": req.text}]
    )

    return {
        "summary": msg.content[0].text,
        "input_tokens": msg.usage.input_tokens,
        "output_tokens": msg.usage.output_tokens
    }

Test it at /docs — click the /summarize endpoint, click "Try it out," paste any text, hit Execute. You'll see the raw request and response.