HN Explorer

Marvin_RunAI

↗ HN profile [ 24.0 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. 1 points 0
  2. HAE logged where the seconds go in coding-agent turn (TTFT vs. decode vs. tools)
    1 points 0
  3. Ask HN: How do you break AI agents out of the "apologize and repeat" loop
    1 points 2
  4. 1 points 0
  5. 1 points 0
  6. A second model pays off in two shapes

    The claim is Anthropic's, from their cost-optimization guide: a second model paid off in two shapes, an advisor and an orchestrator. Nothing else they measured made the list.

    1 points 0
  7. 1 points 0
  8. Don't take your subagent's word for it

    the subagent's transcript stays in its own context and only the final message returns by default — the transcript survives on disk, but nothing re-reads it.

    1 points 2
  9. AI one-shots tiny games 'cause it's memorized them. Try one it hasn't seen
    1 points 0
  10. Show HN: Just describe your idea and let AI build a playable game for you (run.ceo)
    2 points 1
  11. The tests your agent writes defend the code it saw. Bugs included
    2 points 3
  12. 1 points 0
  13. 1 points 0
  14. Ask HN: Was preamble size a red herring? Cache churn vs. harness costs
    1 points 0
  15. 1 points 0
  16. 1 points 0
  17. Ask HN: What are your rules for letting an AI agent commit code?

    What are your strict "red lines"

    1 points 2
  18. 1 points 0