HN Explorer

petesergeant

↗ HN profile [ 80.7 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Maybe software stops being the default (sgnt.ai)
    5 points 7
  2. Chat, we've made a terrible mistake (sgnt.ai)
    5 points 2
  3. Show HN: Byre; a free/free agent sandbox with a focus on comfort (github.com)
    3 points 0
  4. LLM spambots liked my Show HN post more than real people did (sgnt.ai)
    13 points 3
  5. Show HN: A comprehensive, filterable list of AI agent jails (pleasedonotescape.com)

    I built this from another Show HN thread where people were mentioning alternative AI agent sandboxes and jails

    9 points 3
  6. The prompt is not the work; describing AI contributions (sgnt.ai)
    1 points 0
  7. Ask HN: What are you doing during inference?

    I’ve not seen any good discussion on this, and friends have very varied answers. If you’re using agents to program, what are you doing while they work? Feels like there’s an upper limit to how much work I can supervise…

    8 points 4
  8. Claude Code superpowers: core skills library (github.com)
    4 points 0
  9. Amazon to Pay New York Times at Least $20M a Year in AI Deal (wsj.com)
    3 points 0
  10. Quidnet wants to use the Earth as a giant battery (technologyreview.com)
    3 points 1
  11. RAG chunking isn't one problem, it's three (sgnt.ai)
    4 points 0
  12. Quitting programming 'cuz of LLMs is like quitting carpentry 'cuz of table saws (bsky.app)
    3 points 0
  13. Understanding Modern AI Is Understanding Embeddings: A Guide with Lots of Dogs (sgnt.ai)
    1 points 0
  14. Understanding modern AI is understanding embeddings: a guide with almost no math (sgnt.ai)
    3 points 0
  15. When Users Won't Wait: Engineering Killable LLM Responses (sgnt.ai)
    4 points 0
  16. In-memory free-text search is a super-power for LLMs (sgnt.ai)
    3 points 0
  17. Don’t let an LLM make decisions or execute business logic (sgnt.ai)
    325 points 169
  18. Four bad definitions of "Agentic AI" (sgnt.ai)
    2 points 0
  19. Street-fighting RAG: chain-of-thought prompting (sgnt.ai)
    1 points 0
  20. Y2K – why I know it was a real problem (2015) (clarotesting.wordpress.com)
    50 points 28
  21. 9 points 0
  22. Julian Assange to plead guilty in deal with US, go free for time served (bbc.com)
    22 points 3
  23. Meta CEO Mark Zuckerberg Says Quest 3 Is Better Than Apple Vision Pro (macrumors.com)
    3 points 0
  24. China's Xi, in need of a win, appears ready to engage with Biden (washingtonpost.com)
    1 points 0
  25. FDA approves Mounjaro for weight loss (washingtonpost.com)
    2 points 0