Agent Model Fit
Right agent. Right model. Right problem.
Agent Model Fit is a living research and implementation publication for digital and IT leaders deciding where AI belongs in real work.
Start here
Building Agent Model Fit: Right Agent, Right Model, Right Problem
A documented publication build shows how objectives, decision rights, versioned handoffs, and corrections shape agent workâand what the record cannot prove.
Latest articles
All PostsProgrammatic Search Agents: Check What the Model Actually Saw
A practical way to diagnose missing evidence in search-agent answers, informed by Programmatic Search Agents research.
How Much Human Work Did the Agent Need?
A practical log for tracking the clarifications, corrections, and reviews behind an AI-assisted result.
Writing a Correction Notice for an AI-Assisted Business Report
Name the affected report, state the corrected conclusion, tell readers what to do, and preserve the superseded version when an AI-assisted report needs a material correction.
Sections
Explore all sections- Framework
2 Articles
ways to reason about agent, model, and workflow decisions.
- Fieldwork
2 Articles
documented implementation choices, corrections, and lessons.
- Guides
5 Articles
practical procedures a reader can adapt to a defined task.
- Evaluations
2 Articles
how to assess outputs and interpret comparison evidence.
- Business
1 Article
organizational value, readiness, operating costs, and practical tradeoffs.
- Case Studies
0 Articles
bounded accounts of a problem, intervention, and outcome when evidence supports them.