stop playing whack-a-mole with your AI agent. fix your agent for one customer without breaking it for another https://t.co/NPiFjqUCuz 🪞mirrors.dev San FranciscoJoined July 2026
ngl evals are getting less useful with every model release
the better models get at noticing they’re being tested, the less a passing eval tells you about prod
the only environment a model can’t tell apart from prod is prod. reliability work is shifting from “test before you
ngl evals are getting less useful with every model release
the better models get at noticing they’re being tested, the less a passing eval tells you about prod
the only environment a model can’t tell apart from prod is prod. reliability work is shifting from “test before you ship” to “fix it while it runs”
7 Followers 24 FollowingFounder + CTO @ LessonOrca
https://t.co/vyOQ51VHBR
https://t.co/GRCTzYcLPd
Building the future of tutoring via human + agent collaboration.
45 Followers 270 FollowingCo-owner @hireguava. Founder @ SunLyte Productions. Films, growth marketing, and turning long conversations into short content.
38K Followers 2K FollowingI share insights about AI and data science. Sometimes business and finance too.
Your only limit is your mindset.
DM for collaborations ✉️
2K Followers 3K FollowingHelping startups win (DMs open). Founded Pong; founding team @getcallin, @tezos. @Stanford. Love science & technology, the arts, and history.
1K Followers 7K Followinghead of marketing at (redacted) & hanging out with @shefiorg @boysclubworld @berabaddies @miberamaker333 🧸💖 also known for speaking to the stars 🌌
1K Followers 1K FollowingDad. Founder/CTO @hellogumbox. We help businesses build production AI systems. Building and launching software products for 20+ years. Sapere Aude
465 Followers 7K Following"Empty your mind. Be formless, shapeless like water. Pour into cup, it becomes cup; bottle, bottle; teapot, teapot. Water can flow or crash. Be water, my friend
486 Followers 714 Followingco-founder & CEO @mirrorsdotdev (YC F26). fix agent failures without creating new ones
4x published AI/ML research @ NeurIPS, EMNLP