astra playing NetHack: - score: 48,978 - BALROG progress: 49.4% - max dungeon depth: 17 - max experience level: 14 important note: this is a cherry-picked episode (best of 10), and Astra has been able to modify its own game-playing harness no privileged information or unfair game modifications, but it uses tools such as pathfinding to previously visited landmarks and code-written trackers for previously observed entities. these offload a fair amount of low-level reasoning, while Astra remains responsible for choosing what to do at every step imo, the ability of LLMs to autonomously create tools that amortize low-level control costs (and then interpret and use their outputs in-context) is remarkable, and one of the reasons i'm excited about LLM agents. like, we could run the same tool computations and... concatenate the outputs into a flat vector for an MLP... but it would still have to learn what those outputs mean and how to use them. with LLMs we get free hierarchies and natural separation of responsibilities... anyway, i'm now running episodes with the default settings from BALROG for a more fair comparison the auto-research programme continues. will report back! full run below ⤵️

Silico, the platform for ambitious AI research, is publicly available today. AI is advancing fast. The tools to understand it need to advance even faster. Silico lets you interpret and train your models at frontier scale. Learn more + get access 🧵

The arxiv is now live! arxiv.org/abs/2606.18284

Training a model to generate RL tasks not too hard, not too easy costs many solver runs per task. PROPEL predicts difficulty via a probe on its activations instead, amortizing cost and speeding up generator optimization. New open-ended RL research from @Vmax + @GoodfireAI.

The unix terminal is the natural interface for agents to get work done on a computer but how well can agents actually use unix? Claude Code. Codex. Devin. Every frontier agent ships as a terminal tool. With unix-ctf, Vmax is using setters and solvers to measure Unix competence.

Blog post: vmax.ai/team/populora-co-evo… Paper: arxiv.org/abs/2605.16727

So excited to welcome Geoffrey Bradway as Member of Technical Staff @VmaxAI. Geoffrey is a rare catch. He was an engineer at @GoogleDeepMind, Google for Youtube and also has experience in early stage companies, having been a previous @ycombinator founder and also VP of engineering at @numerai. Fitting the Vmax DNA, he has experience with RL before it was cool (doing RL all the way back in 2014). Outside of work, Geoffrey does some really cool art with robotic drawing machines. Cannot wait to share more about what he is cooking

@VmaxAI is excited to have @creus_roger joining us as a research fellow! Roger is joining us from @Mila_Quebec where he works with @pcastr and @GlenBerseth. Roger Creus Castanyer is a brilliant RL researcher working on exploration, credit assignment, and skill discovery. He is also fresh off of a NeurIPS spotlight and a recently accepted paper to ICLR, you can find more of his research in the comments. Roger is significantly accelerating our research on automated environment design - looking forward to sharing what he is cooking!