IDEAS, DECISIONS & LESSONS

Journal

Notes on AI, architecture and building useful products.

32 articles

22 Feb 2026 · 22 min read

The Voice AI Interview Playbook: Building the Voice Pipeline — STT, LLM, and TTS That Actually Work (Part 4 of 12)

Part 4 of 12 — Deepgram Nova-3 or Whisper for STT? Gemini Flash or GPT-4o for conversation? ElevenLabs or Cartesia for TTS? Real benchmarks, real costs, and the combinations that actually work in production.

21 Feb 2026 · 18 min read

livekit

The Voice AI Interview Playbook: LiveKit vs. Pipecat vs. Direct — Picking Your Framework (Part 3 of 12)

Part 3 of 12 — LiveKit gives you WebRTC infrastructure. Pipecat gives you pipeline flexibility. Direct integration gives you simplicity. Here's how to choose for your voice interview platform.

20 Feb 2026 · 29 min read

gemini-live

Building a Production Voice AI Interview Agent: S2S-Only with Dynamic Prompts

A practical build guide for a speech-to-speech interview agent using LiveKit MultimodalAgent, OpenAI Realtime, and Gemini Live. Dynamic system prompts, 3 personas, function calling, and provider switching — no cascaded pipeline needed.

20 Feb 2026 · 11 min read

The Voice AI Interview Playbook: Cascaded vs. Speech-to-Speech — Choosing Your Pipeline Architecture (Part 2 of 12)

Part 2 of 12 — The cascaded STT→LLM→TTS pipeline gives you control. Speech-to-speech models give you speed. Here's how to choose — and why the best systems use both.

19 Feb 2026 · 10 min read

The Voice AI Interview Playbook: Why Real-Time Voice Changes Everything (Part 1 of 12)

Part 1 of 12 — The landscape of real-time voice AI has shifted. Gemini Live, OpenAI Realtime, Bedrock Nova Sonic, and Grok make sub-500ms AI conversations possible. Here's the reference architecture for building a production voice interview platform.