I build human-quality Voice AI systems for real business phone calls. Currently building Goodcall's 4th-generation LLM voice agent with the team — real-time speech pipelines, model orchestration, and production infrastructure on GCP.
An LLM-first rebuild. Real-time voice on LiveKit/WebRTC, ASR → LLM → TTS pipeline, GPT-4 and Gemini with routing and fallback. Thousands of calls per day across hundreds of US businesses. 10+ years shipping software, the last 5 in voice AI.
Open to senior IC rolesYesterday LiveKit shipped Agents 1.5.9. Buried in the release notes, between a Deepgram redaction parameter and a DTMF fix, is the line that matters for anyone running outbound voice agents: Answering Machine Detection (AMD) is now built…
Notes from production — voice AI, TypeScript, databases, and the occasional Node deep-dive.
