nblintao

4 ms·

1. 1.▲
   Miles v0.1: Production-level Post-training(lmsys.org)

   2 points by nblintao 1mo ago|0 comments
2. 2.▲
   No H100? No Problem. Why I Built ML Patron for Autonomous AI Research(medium.com)

   2 points by nblintao 6mo ago|0 comments
3. 3.▲
   An AI Reads the Web: A Deep Dive into Claude Code's WebFetchTool(medium.com)

   2 points by nblintao 6mo ago|0 comments
4. 4.▲
   by nblintao 6mo ago

   Doesn't look like just a thin wrapper to me. The interesting part seems to be the surrounding harness/workflow layer rather than only the model call itself. I was trying to keep track of the better post-leak code-analysis links on
5. 5.▲
   A curated list of high-signal analyses from the Claude Code source leak(github.com)

   1 points by nblintao 6mo ago|0 comments
6. 6.▲
   by nblintao 6mo ago

   The part I keep going back and forth on is openness. This whole model assumes experiments, notes, and results are public. That works great for "open research for the love of it". But even in academia people worry about getting sco
7. 7.▲
   by nblintao 6mo ago

   The "price of an entire new employee" framing is spot on. I kept running into the same thing: individual experiments are cheap, but they add up fast, and nobody wants to approve that budget for speculative ideas. I've been th
8. 8.▲
   Show HN: ML Patron – Run reproducible ML experiments with integrated funding(mlpatron.com)

   2 points by nblintao 6mo ago|1 comments