HNHacker News
TopNewBestAskShowJobs

rad-b

107 karma · joined April 14, 2026

submissionscomments
rad-b··on Aihumanbench
Seems interesting but testing myself only yields my results? How would I compare the result to a frontier model, that part seems to be missing?

Also, the tests seem to be heavily skewed in favor of what LLMs are good at.

rad-b··on How Not to Make an App That Only Has 1% Shot at Turning Profit?
Sounds promising. How do you find ideal customers? That's usually the hard part.
← PreviousPage 2 of 2