开启时会模糊预览图，关闭后正常显示
RT @dbreunig: The evolution of @LiteLLM's model schema illustrates the significant changes in applied AI over 2025. A good chunk of builde…
Asst professor @MIT EECS & CSAIL (@nlp_mit). Author of https://t.co/VgyLxl0oa1 and https://t.co/ZZaSzaRaZ7 (@DSPyOSS). Prev: CS PhD @StanfordNLP. Research @Databricks.
RT @MustafaShukor1: VL-JEPA is out! A non-generative vision-language model, based on JEPA. Different from typical data-space autoregressive…
Distributed Learning @ deepmind | DiLoCo, DiPaCo. Continual Learning PhD @ Sorbonne
This is so bittersweet. We can't really compete in the 6th generation on our own, not enough talent, not enough industry, not enough funding. Isolated, but beautiful, technical flexes to compensate. Very Soviet-like.
We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1
harnesses are an incredibly important piece of agent engineering this makes it really easy to take the world’s most popular agent harness (Claude Code) and see exactly how different models perform wildly differently with different harnesses HarnessBench 👀
agents, harnesses, and evals @LangChainAI, prev @awscloud, phd cs @ temple
RT @thejustinwelsh: Somebody out there is Googling "how to start a business" for the third year in a row. Somebody else started one last m…
The $10M Solopreneur | Helping 100,000+ experts turn their expertise into income.
harnesses and agents are tightly coupled, can’t eval one without the other and harnesses change agent perf A LOT todays harnesses are full on products & companies and the work to get them right is painstaking and requires lots of testing and evals and engineering, so makes sense why it’s such a big business harness engineering has been in 👀
agents, harnesses, and evals @LangChainAI, prev @awscloud, phd cs @ temple