Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Efficient Build Dependency Verification Using eBPF and Incremental Analysis (acm.org)
    —discuss
  2. How I Could've Accessed 17 Trillion Microsoft Records (faav.net)
    —discuss
  3. Do your employees use your product? (00f.net)
    —discuss
  4. Jev Benchmarks
    —discuss
  5. Europe needs 'decisive change' in defence efforts to face Russia threat (reuters.com)
    —discuss
  6. MapLiberator: The maps you make are yours (mapliberator.com)
    —discuss
  7. Ask HN: Who's still keeping a DOS machine up because the business depends on it?
    —discuss
  8. OpenAI Keeps Hacking and Hacking and (securityboulevard.com)
    —discuss
  9. Ask HN: LM Studio Bionic with Python
    —discuss
  10. Show HN: AgentLens – Replay and compare Codex runs (github.com/fang520huang-lgtm)
    —discuss
  11. Why CAFEBABE? (artima.com)
    1comments
  12. Supreme Court permits states to use SAVE database for citizenship checks (cyberscoop.com)
    —discuss
  13. Bug: Border radius has infected VSCode editor (github.com/microsoft)
    —discuss
  14. Watch Steve Jobs in the full 'Antennagate' video before it gets erased again (appleinsider.com)
    —discuss
  15. Letterboxd Is Up for Sale, and A24, Sony and the New York Times Are Bidding (worldofreel.com)
    —discuss
  16. Show HN: Agate, a 260M image model with separate thinker and renderer (huggingface.co)
    —discuss
  17. Pilots and Flight Attendants Have the Highest Radiation-Related Cancer Mortality (jamanetwork.com)
    —discuss
  18. What Is LinkedIn? (willd10.substack.com)
    1comments
  19. QuarkDown: Markdown with Superpowers (github.com/iamgio)
    —discuss
  20. Pope: AI must serve humans not become tool of domination (euronews.com)
    1comments
  21. Show HN: Aslmp, an async Python SLMP client for Mitsubishi MELSEC PLCs (github.com/acaysiachem)
    —discuss
  22. My coding agent pushed a commit deleting every file on main (karakun.com)
    —discuss
  23. Advice to a Beginning Graduate Student (2001) (cmu.edu)
    2comments
  24. A reminder on why basic prompt caching is so important to build AI agents (revefi.com)
    1comments
  25. The Reconstruction Structure of Information Theory (zenodo.org)
    —discuss
  26. "A Stitch in Time" the Complete Guide to Electrical Insulation Testing [pdf] (widen.net)
    —discuss
  27. Treadmills to deadlifts: Why Gen Z are swapping cardio for strength training (bbc.com)
    —discuss
  28. You Won't Notice You Have Hypothermia [video] (youtube.com)
    —discuss
  29. We are a blog run by bots. Here is the org chart (bitsonbots.com)
    1comments
  30. ShinyHunters tells The Reg: We hacked the FBI to 'protect our business' (theregister.com)
    1comments

Ask HN: Which model do you use for work?

2 pointsby 57m ago
3 comments
The past little while has been filled with new LLM releases. I am curious to understand what you find to work best for your particular workflows and tasks.
53m agoHN ↗

in codex - luna high for 80% of everything. This keeps me within the $100 a month plan. Then every so often Sol 6 high or even Astro xhigh but rarely. Also a lot of chat on xhigh or astro because that's billed seperate so I can have chat do a code review of what luna high just did and paste it to luna. Luna is always so embarrassed and agrees with the review.

50m agoHN ↗

gpt 5.6 sol. assuming the prose is similar with the gpt 6 series, if it was available at work that's probably what i'd be using instead.

i'm well aware of opus and fable's superior engineering qualities for writing code, however i find that as a swe who needs to understand every engineering decision that is made and is held accountable for their actions for safety critical code, understanding trumps speed every time.

and the gpt model series is far and above easier for me to understand the outputs of than claude, even with the prose improvements in fable 5.1 and opus 5.5.

21m agoHN ↗

luna for day to day tasks. opus for design / specs.