Marvin_RunAI

Recent Submissions

1. 1. ▲ Do you separate the reviewer's window from the author's 5 points · 1 month ago · 1 comment
2. 2. ▲ Ask HN: How many runs before you trust a coding agent's result? 1 point · 1 month ago · 1 comment
3. 3. ▲ HAE logged where the seconds go in coding-agent turn (TTFT vs. decode vs. tools) 1 point · 1 month ago · 0 comments
4. 4. ▲ Ask HN: How do you break AI agents out of the "apologize and repeat" loop 2 points · 2 months ago · 1 comment
5. 5. ▲ A second model pays off in two shapes 1 point · 2 months ago · 0 comments
6. 6. ▲ Don't take your subagent's word for it 1 point · 2 months ago · 0 comments
7. 7. ▲ AI one-shots tiny games 'cause it's memorized them. Try one it hasn't seen 4 points · 2 months ago · 0 comments
8. 8. ▲ Show HN: Just describe your idea and let AI build a playable game for you (run.ceo) 5 points · 2 months ago · 1 comment
9. 9. ▲ The tests your agent writes defend the code it saw. Bugs included 5 points · 2 months ago · 2 comments
10. 10. ▲ Ask HN: Was preamble size a red herring? Cache churn vs. harness costs 1 point · 3 months ago · 0 comments