HN Explorer

alex000kim

↗ HN profile [ 52.7 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Show HN: NanoRL – RL training for LLMs in ~1,800 lines (github.com)

    The smallest async RL trainer I could write: one loop that runs REINFORCE on CartPole on a laptop and async GRPO on a cluster (e.g. 8xH100 trainer, 8 vLLM workers, ran as a [SkyPilot job…

    11 points 1
  2. Scheduling jobs across Slurm clusters (and K8s, and cloud) from one place (skypilot.ai)
    2 points 0
  3. RL Is Bottlenecked by Inference. Scale It Independently (skypilot.ai)
    11 points 2
  4. Silverback Imfura took a chance, and ended up alone (gorillafund.org)
    75 points 7
  5. Do AI Detectors Work Well Enough to Trust? (chicagobooth.edu)
    5 points 1
  6. You've been doing harness engineering all along (alex000kim.com)
    7 points 0
  7. Lessons from Going Solo (alex000kim.com)
    7 points 1
  8. The Claude Code Source Leak: fake tools, frustration regexes, undercover mode (alex000kim.com)

    Related ongoing thread: Claude Code's source code has been leaked via a map file in their NPM registry - https://news.ycombinator.com/item?id=47584540 Also related: https://www.ccleaks.com

    1376 points 117
  9. 1 points 0
  10. 1 points 0
  11. 1 points 0
  12. 1 points 0
  13. 1 points 0
  14. Bulk Object Storage data migration with SkyPilot (nebius.com)
    5 points 1
  15. ML experiments in the cloud with Skypilot and DVC (alex000kim.com)
    3 points 0
  16. Why Sales Engineers Exist (alex000kim.com)
    2 points 0
  17. Don’t know what to do next? Teach (alex000kim.com)
    2 points 0
  18. Building your Data Science career (youtube.com)
    1 points 0