GernalistYou ask. We investigate. The evidence is public.

Community

AI Builders

Engineers and founders shipping with AI, comparing what models actually do against what the launch posts claim.

Join the investigation

Joining lets you post in the lounge, propose questions, and vote on nominations.

Free · no spam · your name appears on your votes.

Reading is open to everyone. Joining is account-gated.
≈ 4
members
4
active topics
1
on the propose board

Lounge

Casual chat — promote a sharp question to the propose board
  1. Founding deskseed≈ launch seed

    This lounge is seeded by the founding desk. Real conversation starts once members join and sign in to post — nothing here is generated chatter.

  2. Founding deskseed≈ launch seed

    Open thread to kick things off: are AI coding agents replacing junior developers, or just changing the work? If a question sharpens here, promote it to the propose board.

  3. Founding deskseed≈ launch seed

    ‘What even counts as an AI agent?’ came up enough that it became a nomination — see the propose board for where it stands.

Join the investigation

Post in the lounge once you join — real names, no anonymous chatter.

Free · no spam · your name appears on your votes.

Posting opens once you join and sign in. Nothing in this lounge is auto-generated.

Propose

Propose

Send a question to the public evidence desk

Proposals create or support a live question. Optional sources enter human moderation first; no research run starts here.

Submitting joins the selected desk, proposes the question, and records one support vote for your account.

Desk: AI Builders

Votes and new nominations are account-gated. Enough votes send a question to the global desk for selection.

Topics

researching

Are AI coding agents actually replacing junior developers, or just changing the work?Research pass underway; skeptic pass and the community evidence window are still ahead.Are AI benchmarks becoming less useful for buyers deciding what model to use?Research pass underway; skeptic pass and the community evidence window are still ahead.

published

Is the ‘open-source AI is catching up’ claim true for frontier models?Dossier live with claim-level sources and a correction path.