Versalist blog

Notes on testing and improving AI agents

Featured

Teal line graph with a small hand-drawn person checking three labeled points
Run contractsLeaderboards
Jul 4, 20264 min read

A run contract pins down the model, who paid for the run, and who judged it — before the run starts, not after someone asks.

Explore challengesEvaluation guideChallenge docs
Read article

Latest writing

Teal line graph with a small hand-drawn person reviewing experiment points
AutoresearchPrompt optimization
Mar 11, 20265 min read

We got tired of tweaking prompts by hand and calling it optimization

Autoresearcher runs an agent through skill experiments against a rubric you write, and only keeps the changes that actually score better.

Explore challengesRead the CLI docs
Read article
Teal line graph with a small hand-drawn person working beside a terminal window
CLIMCP
Mar 7, 20263 min read

Your agent shouldn't have to leave the terminal to run a challenge

The Versalist CLI and MCP server bring challenges into the terminal and editor, so nobody's copy-pasting context between a browser tab and a repo.

CLI documentationView challenges
Read article
Teal line graph with a small hand-drawn person connecting a skill to a score
EpisodesEvaluation
Feb 28, 20265 min read

A skill can't get better if you can't see why it missed

An Episode runs a skill against a challenge and hands back a repeatable record: what it did, how it scored, and where exactly it fell short.

Run a challengeEvaluation guide
Read article
Teal line graph with a small hand-drawn person marking several rubric dimensions
AI evaluationMulti-agent systems
Jan 19, 20254 min read

Pass or fail hides the exact thing you needed to know

A support agent can route a ticket correctly and still write a reply that's unsafe or useless. A single score can't tell those two failures apart — a rubric can.

Evaluation guideChallenge docs
Read article
Teal line graph with a small hand-drawn person reviewing an execution trace
Meta-reasoningLLM workflows
Dec 26, 20256 min read

You can improve a workflow without ever seeing inside the model's head

We don't have access to what a model is "actually thinking," and we're not going to pretend we do. What we can measure is execution metadata and output quality — and that's enough to work with.

Agent evaluation guidePrompt guide
Read article
Teal line graph with a small hand-drawn person looking across challenge points
AI challengesImpact
Jun 15, 20245 min read

A high score doesn't tell you what a system actually learned

The leaderboard shows you who scored highest. It doesn't show you what the task was really testing, what the score is quietly ignoring, or whether the result holds up under pressure.

Explore challengesAbout Versalist
Read article