OpenAI engineer ships Japan's open-source Stack-chan robot as an official ESP32 SDK example
pbbakkum · x · 2026-11-21
Peter Bakkum, who leads OpenAI's multimodal API development, implemented Stack-chan — the Japanese open-source desktop robot — as an experimental example in OpenAI's openai-live-esp32 repo, clearly marked as unsupported.
- Targets the commercial M5Stack Stack-chan K151/K151-R: ESP32-S3, 320×240 touch LCD, ES7210 mics, GC0308 camera, dual-servo base.
- Adds live voice conversation, voice-directed pan/tilt, and camera Q&A via the GPT-Live-1 Live SDK; a later PR adds 'Cinema' semantic search over camera imagery on-device.
- README documents a full build path: ESP-IDF 5.5.2, idf-component-manager 2.5.2, Wi-Fi profiles via wifi.json (1–8 networks).
- Timeline: GPT-Live-1 API launched 9/10, ESP32 SDK first commit 10/27, Stack-chan support and Cinema landed 11/6 and 11/17 as PRs #30/#31 — an interesting case of a Japanese OSS robot feeding into OpenAI's voice AI experiments, though no official collaboration exists.
More from coding & agent
- Anthropic launches 532 free Claude plugins from Stripe, Shopify, Canva and more — nikola_mr64990 · 2026-11-21
- PostHog cuts LLM costs with decision models: up to 444.6x cheaper, 193.6x faster — yawnxyz · 2026-11-21
- Dev builds GURU, a local-first personal AI framework with persistent memory and device control — iamth3morningstar · 2026-11-21
- Security researcher's 'software factory' playbook: write specs for days, add PR review late — nptacek · 2026-11-21
- Vibe coding with Claude over live voice to build and iterate a portfolio site — Different_Scholar_74 · 2026-11-21
- A 15-Lesson Free Claude Guide to Learn It Over One Weekend — rubenhassid · 2026-11-21