How to Integrate OpenAI into a Django or FastAPI Backend
- Author
- Vishal Maurya
- Published on
- Reading time
- 2 min read
Overview
Adding a model call to a backend is straightforward; making it safe to operate under real traffic requires request limits, timeouts, error handling, and control over what data is sent. Keep the provider integration behind a small service layer so business code is not tied to a specific SDK call.
Define the task, expected output, maximum input size, latency budget, and fallback behavior. Use structured outputs when downstream code expects fields, and validate the result even when the provider constrains the format.
Keep the integration maintainable
- Store the API key in server-side configuration or a secret manager.
- Set explicit timeouts, retry limits, and request-size limits.
- Validate user input and avoid sending unnecessary personal or confidential data.
- Treat model output as untrusted until validated for the task.
- Record latency, token usage, errors, and cost without logging sensitive content.
- Test provider failures, rate limits, malformed inputs, and fallback behavior.
Example: a concrete starting point
import os
from openai import OpenAI
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"], timeout=20.0, max_retries=2)
def summarize_text(text: str) -> str:
if not text.strip() or len(text) > 40_000:
raise ValueError("Text is empty or exceeds the configured limit")
response = client.responses.create(
model=os.environ["OPENAI_MODEL"],
input=f"Summarize the following text concisely:\n{text}",
)
return response.output_text
The example illustrates the boundary and validation pattern; adapt names, persistence, error handling, and provider-specific details to your application. It is not a complete production system by itself.
Failure cases to test
Never expose the API key in frontend variables. Retry only when the operation is safe to repeat, and understand the SDK’s retry behavior. Model names and API capabilities change, so verify current provider documentation before deployment.
Before release, test the happy path, invalid input, missing permissions, dependency failure, and a retry or duplicate request where relevant. Record expected behavior so the implementation can be reviewed against the real business requirement.
When to bring in a developer
I can integrate model APIs into a Django or FastAPI service with validation, observability, and sensible operational limits.
Conclusion
The reliable version of this feature is not just the happy-path demo. It makes ownership, validation, authorization, failure handling, and operational behavior explicit. Start with the smallest useful workflow, measure it against real data, and expand only when the next requirement is clear.