When designing APIs, ensure responses adhere to HTTP semantics by:

1. Using semantic status code constants instead of magic numbers ```python

   Bad

   raise HTTPException(status_code=400, detail=”Inactive user”)

Good

raise HTTPException( status_code=status.HTTP_400_BAD_REQUEST, detail=”Inactive user” )

2. Choosing status codes based on their semantic meaning:
   - 400: Client error (malformed request)
   - 403: Authentication succeeded but authorization failed
   - 204: Success with no content

3. Matching response types to HTTP methods:
```python
# Bad for HEAD requests
def head_item(item_id: str):
    return JSONResponse(None, headers={"x-item-id": item_id})

# Good for HEAD requests
def head_item(item_id: str):
    return Response(headers={"x-item-id": item_id})

1. Testing responses thoroughly, including both status code and content:

   # More complete test
   def test_no_content():
    response = client.get("/no-content")
    assert response.status_code == http.HTTPStatus.NO_CONTENT
    assert not response.content
   

Following these practices improves API reliability, interoperability, and compliance with HTTP standards.