Ex-academic turned Meta researcher on building browser-use agents for personal superintelligence
shuyanzh36 · x · 2026-09-19
shuyanzh36 shares that they left academia earlier this year to join Meta's personal superintelligence effort, focused on making models genuinely good at browser use.
Quoting EdwardSun0909: computer use has been a core agentic capability of Muse Spark since MS 1.1, with the model trained end-to-end to automatically decide between scripts and clicks to reduce latency. The author adds that a 2020 PhD statement-of-purpose dream—building a product their mom could use—has now come true.
More from coding & agent
- Why LLM-Written Docs Fail: Models Optimize for Punchiness, Not Readability — mervenoyann · 2026-11-04
- OpenAI DevDay demo shows forking agent task history at any step — paw_lean · 2026-11-04
- Veteran game dev: "AI builds a full game in a day" posts are mostly hype — bendee983 · 2026-11-04
- Dev builds interactive Excel-to-dashboard with ChatGPT, shares the prompt — TermRemarkable665 · 2026-11-04
- Agents cut dev time from weeks to a day; bottleneck moved to specs and review — Suspicious_Orchid770 · 2026-11-04
- Orgo, a YC-backed platform giving AI agents their own cloud desktops, claims 1,000+ teams — Scobleizer · 2026-11-04