Top all-time · #15000
Abhishek Harshvardhan Mishra
Personal blog · tokenbender.com/postsaiprogramming
Abhishek Harshvardhan Mishra writes tokenbender, a personal blog at tokenbender.com about ai and programming. Since 2025, 5 different Hacker News readers have linked it 5 times, most often “Avatarl: Training language models from scratch with pure reinforcement learning”. No feed was found, so read it on the site itself.
Abhishek Harshvardhan Mishra works on post-training and autoresearch metaharnesses.
Most-linked posts?
- Avatarl: Training language models from scratch with pure reinforcement learning4 people · 4 mentions · first linked 2025-08-09
What Hacker News says?
Submitted Avatarl: Training language models from scratch with pure reinforcement learning as a story · 9 points
HN discussionSubmitted Avatarl: Training language models from scratch with pure reinforcement learning as a story · 2 points
HN discussionSubmitted Avatarl: Training langauge models from scratch with pure RL as a story · 2 points
HN discussionSubmitted Avatarl: Training language models from scratch with pure reinforcement learning as a story · 2 points
HN discussionSubmitted Honey, I Shrunk the Circuits as a story · 1 points
HN discussion