Every launches Checks, a personal benchmark platform for rating AI models on your real work
every · x · 2026-10-08
Every has built Checks, a personal benchmarking platform that lets anyone measure how good new frontier models actually are at their real work, and is hiring someone to lead the product.
- The idea came from Dan Shipper's brief to quantify the team's "vibe checks" of new models
- 15 internal users have built personal benchmarks so far, including ops lead Arielle Shipper and Kieran Klaassen (of Compound Engineering)
- The team includes people who built workflows for hedge funds, the Cora early team, and an internal AI editor clone (KatePass) of their editor-in-chief
Related event: Every Launches Personal Benchmarking Platform Checks(3 posts)→
More from Apps
- More visual answers: OpenAI shows off GPT-6 Intelligent UI — OpenAI · 2026-10-09
- Play against GPT-6: OpenAI demos Intelligent UI with an in-chat game — OpenAI · 2026-10-09
- OpenAI rolls out GPT-6 Intelligent UI in ChatGPT, generating interactive UIs for every answer — OpenAI · 2026-10-09
- X's new subscription bundles Grok and Cursor with one shared usage pool — nima_owji · 2026-10-09
- Ferryman: cross-posting tool targeting $10,000 MRR, $30-100/mo tiers — KevinNaughtonJr · 2026-10-09
- Google and Unity launch AI playground that builds games from text, plus Spark — glenbeer · 2026-10-09