PM OS vs Team OS vs Company OS: how to layer your AI setup and which to build first
aakashgupta · x · 2026-10-21
Aakash Gupta breaks company AI adoption into three layers, drawing on practices from two CPOs (Jiaona Zhang at Laurel, Mikhail Shcheglov who made his company reachable via Telegram):
- PM OS (personal): your own Claude Code setup holding your context, voice and shortcuts; Zapier CEO Wade Foster rates it "Adoptive."
- Team OS: a shared GitHub repo in Claude Code holding team calls, decisions and working notes, so the same few people stop answering the same questions.
- Company OS: company-wide skills and context across Slack, Telegram, iMessage and GitHub, reading transcripts and email, with human-approved changes and a security/permission layer; rated "Transformative."
The separator: who needs the answer — only you (PM OS), your team (Team OS), or someone after you leave (Company OS).
Which first: flooded with context pings → Company OS; small team on one codebase → Team OS; new to Claude Code → start with PM OS.
His team's Company OS passed a real test: asked for a media kit, the bot read 12 months of sponsor email, found three sponsors who'd asked, and produced a reviewed PRD and a live page after one Slack message.
Related event: Building PM OS, Team OS and Company OS: AI Work Systems in Practice(2 posts)→
More from coding & agent
- LangChain's Managed Deep Agents Bet on Model Independence Over Rivals — hwchase17 · 2026-10-21
- Burkov: without explicit checklists, coding agents hand you 80%-done work — burkov · 2026-10-21
- Microsoft's Self-HPO: Coding Agent Redesigns Its Own Harness, +7 Points on SWE-bench — mark_k · 2026-10-21
- Laude Institute launches Slingshots FOUR: 14 projects across 8 top institutions — a1zhang · 2026-10-21
- Research agent Épi sets 120 records, narrows n-queens constant interval 84x — dair_ai · 2026-10-21
- Compaction hooks in practice: keep last n messages verbatim, summarize the rest for full agent continuity — MoonL88537 · 2026-10-21