Live data from Hacker News

Viewing profile — nravic

nravic

HN member
Joined
Mon, May 01, 2017, 3:42 PM UTC
HN karma
28
Public activity
43 items

About nravic

[ my public key: https://keybase.io/nravic; my proof: https://keybase.io/nravic/sigs/-W7C_1M4aYXWPyuN8wQdlxtuDKdPESuSpqvyZRykcTc ]

Recent public activity

  1. comment
    Comment #45014612

    I think even easier in fact - what's happening behind the scenes w/ an LLM is far more opaque

  2. comment
    Comment #44327062

    This is super interesting! We do something similar I think by taking a checkpoint after model initialization. I'm curious what you think about our approach, here's some benchmarks:…

  3. comment
    Comment #40755100

    How were you handling GPU state w/ pytorch? We added some custom code around CRIU to enable GPU checkpointing fwiw: https://docs.cedana.ai/setup/gpu-checkpointing/

  4. comment
    Comment #40753702

    We do this using CRIU right now! https://github.com/cedana/cedana In fact one of our customer's use cases is exactly what you describe, allowing users to "hibernate" container work…

  5. comment
    Comment #40753680

    I'm probably the cofounder of the guy you spoke with! Here's our repo: https://github.com/cedana/cedana

  6. comment
    Comment #38305917

    what kind of pannier bags do you use to haul groceries? All the ones I've used have been too small for my purposes.

  7. comment
    Comment #37226709

    IIRC (and in my experience) KL divergence doesn't account for double counting. Wrote a paper where I ended up having to use a custom metric instead: https://digitalcommons.usu.edu/…

  8. comment
    Comment #37189756

    This is so cool! Out of curiosity, could you walk through your decisionmaking around using Nix and how it fits in with Rivet?

  9. comment
    Comment #37079124

    how often are these batch jobs run? I'm curious to know what the absolute maximum sync frequency can be.

  10. comment
    Comment #36871354

    Self plug: run llama.cpp as an inference server on a spot instance anywhere: https://cedana.readthedocs.io/en/latest/examples.html#runnin...

  11. comment
    Comment #34563830

    GrapheneOS uses WebUSB to flash your phone. Pretty wild first experience doing it for me - I'm used to the hassle of sideloading a bootloader and then flashing.

  12. comment
    Comment #32014363

    Hey! I'd love to learn more about the position - the hook @ jpl email on your website is bouncing though. Do you mind sending me contact details so we can chat? Thanks :)

  13. comment
    Comment #28727833

    Do people out of college even do Leetcode anymore? Surely there's a better use of time once you're in a paying job.

  14. comment
    Comment #23258967

    If you're looking at aerospace/defense startups, this is usually the case. It's not unheard of for them to employ people who's primary role is grant writing to try and get (for exa…

  15. comment
    Comment #22171208

    Zotero is my go-to

  16. comment
    Comment #22141363

    Julia's REPL is beautiful

  17. comment
    Comment #22065818

    Yeah, spiking neural networks are a tough nut to crack. Check out Nengo if you're interested in learning more

  18. comment
    Comment #22047615

    What are Protonmail's problems?

  19. comment
    Comment #21649851

    very visceral reaction there

  20. comment
    Comment #21615290

    I've followed the same path, out of curiosity do you mind if I PM you asking a few questions? I'm a couple months out of undergrad and I'd like to pick your brain a bit.

  21. comment
  22. comment
    Comment #21518327

    This is really cool, what are you training?

  23. comment
    Comment #21466458

    The population argument is fundamentally flawed. It places the burden of climate change onto the global south, which is far from the actual truth

  24. comment
  25. comment
    Comment #21455395

    pleasantly surprised to see julia used in this