Unofficial. A job application for Linear’s Product Manager role, built from public pages. Not affiliated with Linear.

How I built this.

I ran this with Claude as my lead agent. AI agents did the research and built the pages. Here is the workflow, what was checked, and where the AI was wrong.

The workflow

MeSet the approachPitch the role with a plan and a working prototype. Read the company as its user would before pitching anything.
AIRead everything publicOne agent on Linear: 654 pages saved, with 105 documentation pages, 255 changelog entries and the public API schema. A second agent on six rivals.
AICheck each findingEvery claim about Linear was checked against the saved page text. The starting idea did not survive the check.
AIBuildThe proposal, the prototype and a private findings report.
MeRead it and stand behind itI read every page of this site before sending it.

The lead agent was Claude. It ran the research, did the checking and wrote the pages. One working day, November 22, 2026.

The rules the agents worked under

  • Every claim carries its source.A page address and where on the page. The page text is saved, so any claim can be checked again.
  • “Not found” is never rewritten as “missing”.The agents had to say where they looked. Only a statement by Linear itself counts as confirmed absent.
  • Read only.No account, no login, no form, no testing of any system. Nobody opened the Linear app.
  • Separate agents for Linear and for rivals.So what one vendor says about another never stood in for that vendor’s own page.

Where the AI was wrong or thin

  • The starting idea was wrong.The research began by looking for missing cross-team sprint planning. Linear’s sub-teams already share one cycle schedule. The proposal moved to the gap Linear’s own FAQ confirms.
  • One claim could not be settled, so it was cut.An early draft said no screen shows what a closed cycle handed on. A 2020 changelog line suggests the app does show it. Without opening the app there is no way to know.
  • The demand evidence is thin.Linear has no public forum that the agents could read. The case rests on Linear’s own FAQ and one customer story, not on customer requests.
  • The rival rows are thinner than the Linear row.Two help centres could not be read, and the rows were not checked line by line. The evidence page says so.

Where my judgment came in

The agents can read 650 pages in an hour. They cannot decide what a hiring team should be shown.

  • Pitch with a plan and a working prototype.Linear’s hiring page says the best talker is not always the best doer. A page you can use is the doing.
  • Read the company as its user would, before pitching anything.The lens here is my own job: ten vendor teams and a date that does not move.
  • Report what turned up, privately.If I find holes, I bring them up. The findings went into a report, not onto this site.
  • Say what I cannot see.The proposal has a section on what would change it, because I have no view of Linear’s data or its roadmap.
  • Stand behind the output.I read every page of this site before sending it. If something here is wrong, that is mine, not the agent’s.