production build OCT 16, 2026
Your Voice Agent Needs a Better Harness, Not a Better Model
A strong model won't save a voice agent with a weak harness. How I test voice agents: human calls, three tiers of automated tests, and guardrails.
Piotr
Engineering Insights.
Technical deep dives, architectural protocols, and engineering notes on deploying Voice AI at enterprise scale.
production build OCT 16, 2026
Why I Don't Build Voice Agents on Vapi, Retell, ElevenLabs or Bland
Why I build voice agents on LiveKit and Pipecat instead of closed platforms: SDK limits, per-minute pricing, BAA fees, testing and self-hosting.
Piotr
MAY 14, 2026
5 Costly Mistakes I Made Building Voice AI (And How to Fix Them)
A breakdown of the most expensive and time-consuming mistakes made when building production Voice AI agents, and how to avoid them.
CO
cost and quality MAY 14, 2026
Speech-to-Speech, No-Code -> Great Tools or Just Hype?
Why no-code fails, bypassing LLMs to cut 700ms latency, and the limits of speech-to-speech models.
Piotr
compliance MAY 07, 2026
HIPAA and GDPR Compliance in Voice AI: A Practical Guide
How to build compliant Voice AI pipelines without sacrificing performance. A deep dive into PHI/PII redaction, vendor agreements, and architectural trade-offs.
Piotr
cost and quality APR 30, 2026
The Surface Area of Disappointment: Hard Voice AI production lessons from a $3.2M a16z-backed founder
A conversation with Ezra AI Labs' Ophir Samson on the three stages of Voice AI disillusionment, the long tail that breaks demos in production, and why expectation management closes enterprise deals.
Piotr
cost and quality APR 16, 2026
How to Cut Voice AI Costs: Lessons from 50M+ Calls
Practical lessons on telephony billing, verification UX, and when to optimize LLM, TTS, and STT from 50M+ production calls.
Piotr