Benchmark cases

Case catalogue

Every case connects a product baseline to model runs, reports, and linked evidence. Status is operational context, not the identity of the case.

Active internal

CASE

The IdeaGrains website itself is a benchmark case for testing whether AI can reposition, simplify, and structure a product-testing platform.

PositioningInformation architectureHomepage clarityVisual hierarchyBenchmark catalogue structureTrust/disclaimer content

In progress

Active

CASE

A data-heavy chess scouting product used to test whether AI models can improve UX, insight quality, report clarity, and code reliability.

Data-heavy UXInsight rankingReport clarityBackend logic

Published draft

First model comparison

Planned

Planned

CASE

A published browser extension used to test small utility improvement, onboarding, and Chrome/Edge UX.

Browser extension UXTab managementOnboarding

In progress

Revival candidate

CASE

A parked writing product used to test whether AI can diagnose, reposition, rebuild, or recommend retirement.

Product revivalRepositioningStack decisionsMonetisation

Infra note: Database currently offline

Related reports

Can AI revive a parked writing product?

Planned

Which stack would AI choose for WriteAndWait?

Planned