Evo 2 Team Credits a Researcher Spanning RL Post-Training, Inference Optimization, and Neuroscience
BrianHie · x · 2026-10-29
In the closing tweet of the Evo 2 work thread, Brian Hie highlights lab member Alex's rare skill range: RL post-training, inference stack optimization, molecular neuroscience, mouse surgery, and whole-brain single-cell RNA-seq.
Credits go to the core team (garykbrixi, matthewjzliu, and others), Will Allen's lab for scRNA-seq assistance, and Lawrence Livermore for collaborating on lanmodulin design.
More from Research
- CausGround: non-parametric causal grounding enables adaptive embodied planning without retraining — _akhaliq · 2026-10-30
- Evo 2 genomic model now runs on vLLM: 7B fits 1-megabase context on a single H100 — anshulkundaje · 2026-10-30
- Cusp AI releases ODAC25, the largest open dataset for direct air capture materials — BenBlaiszik · 2026-10-30
- Cambridge paper trains LLMs to ask informative questions via information-gain rewards — _akhaliq · 2026-10-30
- NVIDIA offers free browser-based protein design with RFD4-Proteina until Nov 4 — AllThingsApx · 2026-10-30
- Machenta's 'one-person factory': robot assembly line taught in 2 minutes with sim-trained model — chris_j_paxton · 2026-10-30