Pinned post
if there is one thing that you must not do is surrender
don’t surrender your dreams, your passion, your curiosity or your freedom
never
don’t surrender your dreams, your passion, your curiosity or your freedom
never
@yacinelearning
Pinned post
one thing I love to do is sit in random spot I used to sit a year ago and pull up a notebook and look how far I’ve comeView quoted post on X
alright folks if you are in SF next week I'll be presenting a workshop at the AI engineering code summit really hope to see you thereView quoted post on X
we're going to sit down take a deep breath and review what the heck happened in 2026 with ai agents it's going to be a full hour bring snacks
[1/3] turns out you can teach a 35b model to weight-space program better than opus o_0View quoted post on X
earlier this week, @rikiyatakehi and @jayden_teoh_ showed that astra can manually write neural networks to solve atari games, one weight at a time. they call this weight-space programming.
on the @trajectorylabs platform, @o__jerry__o and I post-trained 27b and 35b-a3b qwen models to write networks for atari games. we curated WeightProgramBench, a set of 322 simple games to teach models to manually write neural networks.
we combined 51 atari-style games with 266 games adapted from crisp-game-lib for training, holding out the 5 original minatar games (freeway, breakout, asterix, seaquest, space invaders).
on freeway, qwen-27b climbed from averaging 21.0 to 47.1 points **without ever training on freeway**, but still not quite at astra’s 63.9 points.