HN Explorer

endorphine

↗ HN profile [ 127 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Congress moves to integrate US and Israeli militaries (responsiblestatecraft.org)
    34 points 11
  2. Ask HN: Does Codex hits limits more easily now?

    The past month or so, I have the impression that Codex started hitting the 5h limits much more often. I always used the latest model in xhigh, but would almost never hit any limit (I'm not a very heavy user). Lately,…

    2 points 1
  3. The Oil Market Breaking Point Is Here (hfir.com)
    7 points 0
  4. Token Anxiety (twitter.com)
    3 points 1
  5. Ask HN: Do global AGENTS.md with coding principles make sense?

    For example, greenfield advice like: "prefer clear over clever code. Strive for maintainable, simple and testable code etc". Do these kind of vague advice make sense to have on an AGENTS.md file? And if so, any reason…

    4 points 3
  6. What changed in tech from 2010 to 2020? (tedsanders.com)
    3 points 0
  7. 1 points 0
  8. 19 points 0
  9. Extrepo (manpages.debian.org)
    1 points 0
  10. List of Genocides (en.wikipedia.org)
    9 points 2
  11. List of countries by proven oil reserves (en.wikipedia.org)
    5 points 0
  12. Ask HN: Claude Code vs. Codex vs. GitHub Coding Agent?

    With so many options, I'm looking to settle on one of those solutions to use for work in an existing small-to-medium codebase. My goal is to delegate some tickets (menial boring tasks) to the LLMs. Any preference…

    3 points 1
  13. Ask HN: Are LLMs good for implementing features in existing web apps?

    So I have a typical medium-sized Rails codebase, but with significant code that's not the dead simple CRUD stuff. For example, it talks with third-party APIs, does user authentication, uses Postgres FTS plus some other…

    2 points 1
  14. When More Cores Means Less Speed: Debugging PyTorch with Valgrind on ARM (blog.cloudkernels.net)
    2 points 0
  15. Ask HN: Do you think there's censorship on HN?

    It could just be me, so I'd like to know what everyone thinks: lately there have been significant geopolitical tensions and multiple wars being fought. However, I have a feeling that not all news make it to the top of…

    37 points 37
  16. India–Middle East–Europe Economic Corridor (en.wikipedia.org)
    4 points 0
  17. Zig: First Impressions (deplet.ing)
    7 points 0
  18. 5 points 0
  19. Ask HN: OpenAI models vs. Gemini 2.5 Pro for coding and swe

    In your experience, which of the two models (all of OpenAI vs Gemini 2.5 Pro) are better for having as assistants to ask SWE/software systems related questions and doing long and complex reasoning? I'm debating whether…

    4 points 5
  20. C and C++ prioritize performance over correctness (2023) (research.swtch.com)
    93 points 130
  21. Reporter asks random people "what is a browser?" (2009) [video] (youtube.com)
    1 points 0
  22. Ask HN: How do you pick between different OpenAI LLMs?

    Let's say I'm searching for "how to disable monitor sleep in Linux/Xorg". There are a few options now: 1. Use o1 2. Use 4o 3. Use 4o with web searching enabled 4. Use legacy GPT4 How do you pick between all these knobs?…

    1 points 1
  23. Ask HN: How do you pick between different OpenAI LLMs?

    Let's say I'm searching for "how to disable monitor sleep in Linux/Xorg". There are a few options now: 1. Use o1 2. Use 4o 3. Use 4o with web searching enabled 4. Use legacy GPT4 How do you pick between all these knobs?…

    2 points 0
  24. How Tcmalloc Works (2013) (jamesgolick.com)
    2 points 0
  25. Ask HN: Being Concrete in Tests

    I remember reading in a book that in tests, you generally want to be concrete, not abstract. For example, you'd like to do: assert(response.status == 401) assert(error == "param 'foo' is invalid") instead of:…

    1 points 0