Human evaluation for the machine age

AI writes.
You judge.

Verdict is where verified developers review AI-generated code — scored for accuracy against hidden gold standards, ranked in leagues, and credentialed for it.

How it works

01

Review

Real AI-generated code with real flaws. Score it on correctness, security, readability, and maintainability — then write your critique.

02

Accuracy

Hidden gold-standard items measure your judgment, not your output. Accuracy is the only currency here.

03

Leagues

Climb from Clerk to Chief Justice. Promotion demands proven accuracy — volume alone gets you nowhere.

04

Credential

A public, portable record of your judgment. Top-percentile proof for your README, your LinkedIn, your next role.

Verticals

Every vertical opens with identity you can verify and work you can judge. Developers first.

Developers

Live

Sign in with GitHub. Your commit history is your credential; your reviews become the record.

Artists

Coming soon

Portfolio-verified review of AI imagery, judged by working artists.

Writers

Coming soon

Machine prose, held to a human editorial standard.

“The web fills with machine output while original human signal declines. Models are beginning to train on their own echoes. Verdict manufactures the scarcest resource in AI — structured, credentialed human judgment.”

The thesis