Contact
Problems I Think About
- Scaling WebRTC infrastructure — SFU architecture, cascading, multi-region deployment
- Voice AI latency — building pipelines where the response starts before the user finishes thinking
- Cloud architecture for media — why standard patterns break and what to use instead
- Integrating AI into live communication — real-time STT, LLM, and TTS within WebRTC sessions
If you’re working on something in these areas, I’d enjoy talking about it.
Get in Touch
The best way to reach me for anything substantive:
Schedule a Conversation
If you’d prefer a live conversation, grab a time that works:
Connect
I typically respond within 24-48 hours.