build an OpenAI-compatible multi-provider chat completions proxy

[lukeocodes]

Integration: OpenAI-Compatible Multi-Provider Chat Completions Proxy for Voice Agent API

What this should show

A proxy server that sits between the Deepgram Voice Agent API and multiple LLM backends (Amazon Bedrock Agents, OpenAI). The proxy exposes an OpenAI-compatible chat completions endpoint so the Voice Agent can route requests to different providers via configuration, without changing application code. Should demonstrate:

• OpenAI-compatible /v1/chat/completions endpoint
• Provider switching via config (Amazon Bedrock Agents, OpenAI)
• Integration with Deepgram Voice Agent API as the LLM backend
• Minimal setup to swap providers

Credentials likely needed

• DEEPGRAM_API_KEY
• OPENAI_API_KEY
• AWS credentials (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION) for Bedrock

---

Original request:

What's on your mind?

An OpenAI-compatible multi-provider chat completions proxy for the Voice Agent API, supporting Amazon Bedrock Agents and OpenAI with easy provider switching. Enables developers to swap LLM backends behind a Voice Agent deployment without changing application code.

Any extra context? (optional)

No response

[github-actions]

Ooo, we'll get right on that! 🎉

The Engineer will pick this up shortly and build a multi-provider chat completions proxy for the Deepgram Voice Agent API example — showing how to swap between Amazon Bedrock Agents and OpenAI without changing your app code.

I'll keep this issue open so you can track progress — we'll close it automatically when the PR merges.

If you have any extra context (preferred language, specific API, or credentials you're already using), drop it here — it helps the build go faster!

[github-actions]

Built in #210