Voice agent stack breakdown: Twilio, Deepgram, Cresta and ElevenLabs with a 300ms latency rule
schwentker · x · 2026-10-09
Builder @schwentker shares the full stack behind his voice agent Veronica:
- Phone line: Twilio
- Ears: streaming STT (e.g. Deepgram)
- Brain: Cresta Conductor
- Voice: ElevenLabs
- LiveKit WebRTC can save 300ms of latency
He notes silence feels unnatural past 300ms, and the core challenge is distinguishing a mid-sentence pause from a finished turn.
Next step: having Veronica plan team nights out — 12 SF startup events ingested, plain code applies hard rules ($40 budget, Thursday, fits 8, near SoMa) down to 3 survivors, and a Jev classifier picks only from that filtered list so it can't invent events. "Code filters. Jev picks." Today it serves one person; next, teams.
Related event: Veronica: A Hackathon Voice Agent for Billing Complaints(5 posts)→
More from coding & agent
- Grok Bot capability list spans email, docs, CRM, GitHub and multi-bot workflows — XFreeze · 2026-10-09
- Open-source huashu-art-motion turns coding agents into art-animation studios, 2.6k stars — AlchainHust · 2026-10-09
- Learn marketing engineering with Claude Code and GitHub, marketer argues — FinanceYF5 · 2026-10-09
- Open-source repo ships 49 free Claude marketing skills for Google/Meta Ads and SEO — FinanceYF5 · 2026-10-09
- "Marketing Engineering": 11 drills to build marketing agents with Claude Code and GitHub — FinanceYF5 · 2026-10-09
- Non-coder huashu hits 100k+ GitHub stars by shipping AI-native skills and apps — AlchainHust · 2026-10-09