Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Tracking Singularity – A Curated Log of Important Events in AI Since Mid 2026 (trackingsingularity.com)
    —discuss
  2. SunCalc – sunrise, sunset, shadow length, solar eclipse (suncalc.org)
    —discuss
  3. Picking a domain name is hard (aetherspe.cc)
    —discuss
  4. Show HN: Sitcom Flavour – Season your Claude Code chats with sitcom replies (ada.tools)
    —discuss
  5. Stanford used AI to replace Hispanic student with Black woman in photo (msn.com)
    —discuss
  6. Goodbye Google (ocallahan.org)
    —discuss
  7. Krakatoa, an interactive 2.5D painting of the 1883 eruption (echohive.ai)
    —discuss
  8. Jev Based Code Review (github.com/egma-ai)
    1comments
  9. Plex: Remote Streaming via Third-Party Apps to Require a License (netguide.io)
    —discuss
  10. Proofmode (proofmode.org)
    —discuss
  11. Vulnerability Cve-2026-82958 (circl.lu)
    —discuss
  12. What your phone is made of (anas-sabbar.ca)
    —discuss
  13. Fzakaria/omnibin: Every binary nixpkgs ever shipped, on your PATH (github.com/fzakaria)
    —discuss
  14. Finding Bugs (matklad.github.io)
    —discuss
  15. What's New in Oracle Solaris 11.4 SRU 95 (oracle.com)
    —discuss
  16. Pompeii test case demonstrates Gaussian Splatting's potential (gim-international.com)
    —discuss
  17. Every Package Is Installed (fzakaria.com)
    —discuss
  18. Perch: Semantic Code Linting with Jev (github.com/lakeday-org)
    —discuss
  19. Is indirect prompt injection still a big threat as models get more advanced? (realarcherl.github.io)
    4comments
  20. AI/ML Engineer (docs.google.com)
    —discuss
  21. We Built Safety into Muse (meta.ai)
    —discuss
  22. Micron Technology sucessfully demonstrates first 512GB DDR5 RDIMM module (micron.com)
    —discuss
  23. September's ozone hole biggest in 20 years, warning signs appeared months ago (rnz.co.nz)
    —discuss
  24. Collaborative Editing in Wordgard (marijnhaverbeke.nl)
    1comments
  25. Plan Mode Is Dead (aymannadeem.com)
    —discuss
  26. Power your agents: Gemini 3.8 Live with Live Avatar is now generally available (cloud.google.com)
    —discuss
  27. What I Found Out About Organ Donation Shocked Me (nytimes.com)
    3comments
  28. Scalable Subagents (github.com/ringlochid)
    —discuss
  29. We Built a City of 10k AI Researchers (lab.cloud)
    —discuss
  30. Congestion pricing funds $5M program for overnight truck deliveries in NYC (gothamist.com)
    —discuss

Ask HN: Do you token max? Why or why not?

1 pointsby 3h ago
1 comments
I posit that:

1. I am faster using agentic coding than without

2. I am 10x faster using many agents in parallel

3. When I hit a limit (5h, week), its a good time to take a break - and focus on some slower work or step away.

4. If I can reliably produce quality output with agents, then I should attempt to automate the workflow from idea to QA.

5. If I can automate how I work, I can sleep while I work.

6. If I can sleep when I work, I can get more done, while working less.

7. If I can token max while retaining and continually improving the process and output quality, I can work less and earn more.

3h agoHN ↗

For that there's Python: with30 lines you can automate this task. If you want I can send you the script I use. DM me if interested.