Working on a blog that covers recent research on agents + world modeling. One of those rare topics that is both (relatively) easy to understand and…
RT TechCrunchPrime Intellect raises $130M Series A to help enterprises build their own AI agents https://techcrunch.com/2026/07/08/prime-intellect-raises-130m-series-a-to-help-enterprises-build-their-own-ai-agents/?utm_source=dlvr.it&utm_medium=twitter
RT Craig CertoThis weekend I read @cwolferesearch write up on Agentic RLI highly recommend! I actually printed it out on paper and I enjoyed it so…
Really interesting work. I find it especially interesting that evaluating w.r.t. perplexity can "hide" the damage done by pruning in certain cases. Choosing evals carefully is…
I'll be spending America's 250th birthday in the most American way possible (i.e., binge reading Nemotron / Arcee Trinity tech reports). Happy 4th, all!🇺🇸
Every time I read Sebastian’s books, I understand LLMs better. I can’t recommend his build an LLM / reasoning model from scratch series enough. Reading the…
RT Bhavin JawadeProgress at the AI frontier is bottlenecked by high-quality training data. Meta just dropped Autodata, a framework that uses an AI agent acting as…
Super interesting practical application of KV cache compaction! This is one my favorite areas of research to read about recently. Lots of great papers on the…
Another point to consider that drives home the importance / complexity of environment management:If we assume one environment per rollout in RL, we can have O(100)…
RT Sergio Paniegocompletely agree with this take that the hardest part of agentic RL is managing and scaling the environments, not the algorithm (and why I…