building backend with chat completions api

If you’re building a backend around the Chat Completions API, a good setup usually looks like this:

Basic flow

1. Frontend sends user message to your backend.
2. Backend validates/authenticates the request.
3. Backend calls the Chat Completions API with:
   • a system message
   • conversation history
   • the latest user message
4. Backend returns the assistant response to the frontend.
5. Optionally store the conversation in your database.

Minimal Node.js example

import OpenAI from "openai";
import express from "express";

const app = express();
app.use(express.json());

const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

app.post("/chat", async (req, res) => {
  try {
    const { messages } = req.body; // [{ role: "user", content: "Hello" }]

    const response = await client.chat.completions.create({
      model: "gpt-4o-mini",
      messages,
    });

    res.json({
      reply: response.choices[0].message.content,
    });
  } catch (err) {
    console.error(err);
    res.status(500).json({ error: "Chat request failed" });
  }
});

app.listen(3000, () => console.log("Server running on port 3000"));

Example request body

{
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "Write a haiku about rain." }
  ]
}

Recommended backend practices

• Never expose your API key to the client.
• Keep chat history on the server if you need persistence.
• Trim old messages to avoid token bloat.
• Add rate limiting and auth.
• Handle timeouts/retries gracefully.
• If you want token-by-token UX, use streaming.

Streaming example

const stream = await client.chat.completions.create({
  model: "gpt-4o-mini",
  messages,
  stream: true,
});

for await (const chunk of stream) {
  const token = chunk.choices[0]?.delta?.content || "";
  process.stdout.write(token);
}

Common backend architecture

• Routes: /chat, /conversations/:id
• DB tables:
  • users
  • conversations
  • messages
• Services:
  • openaiService
  • conversationService
  • authService

If you want, I can give you:

1. a full Express backend template
2. a Python/FastAPI version
3. a production-ready architecture with DB + auth + streaming