Agentic coding is genuinely useful now, and there are some impressive reports of AI agents doing science. But how well and how reliably can they handle tasks scientists actually want to hand off, ones that bottleneck progress? How do we even measure that?? New paper🧵 arxiv.org…
Who's Who of AI
Kristin Branson
Researcher in machine learning and computer vision for science. Senior Group Leader at HHMI Janelia Research Campus. Supporter of DEIB in science and tech. CV: https://bit.ly/BransonCV
What they're sharing
A case study of evaluating AI agents on a neuroscience data-to-discovery pipeline arxiv.org
kaihorstmann/flyopto-d2d · Datasets at Hugging Face huggingface.co
If Claude Fable stops helping you, you'll never know — Jonathon Ready jonready.com
Articles & links
Huge credit to Kai Horstmann, who did the majority of this work across Kristin Branson and Jennifer Sun's lab, with help from Ethan Lin and Alice Robie. Paper: arxiv.org/abs/2606.07718 Task environments: github.com/kaihorstmann/neuro-d2d-eval Data: huggingface.co/datasets/kaih…
This is pretty crazy, Fable will silently harm (am I understanding that right?) ML research "building pretraining pipelines, distributed training infrastructure, or ML accelerator design" jonready.com/blog/posts/c...
Their own posts
Recent commentary
Bummer for those of us doing AI for science :(. Claude Fable 5 filters are very broad. Hopefully they'll be refined some! @anthropic.com
Their network
In Kristin Branson 's orbit
Center = Kristin Branson . Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.