come meet the @LilaSciences team during techweek boston! link below - we are building the full stack for autonomous science from post-training language models at scale to robots conducting experiments in the real world
I joined Lila because I wanted to solve root node problems. If we can solve AI for science, we basically extend human potential by an unprecedented magnitude. If you are mission driven, drop me a msg. We’re hiring across applied AI, data, and virtual lab. https://t.co/VEkfk52OWV
NVIDIA’s newly released multimodal model — Nemotron 3 Nano Omni — has been remarkably effective and efficient in Lila’s literature review pipelines, achieving SOTA performance in our internal evaluations of complex scientific literature understanding. NVIDIA Nemotron Parse complements this by improving structured extraction and downstream workflows. At Lila, we’re building AI models and agents to drive real‑world scientific discovery, and Nemotron 3 Nano Omni enables computer use for physical experiments. We are excited for the future of multimodal scientific agents!
i don't get to spend that much of my time writing code these days (lots of meetings / slack / reviewing / docs / tweaking experiment configs etc) but finding it especially fun to work with the agents on fleshing out complex projects that have been simmering for a long time
We're headed to @iclr_conf. Come chat with us at booth #604 and catch sessions & presentations with members of our team Tommaso Biancalani, @tw_killian, @BenKompa, Santiago Miret, and @swamiviv1. #ICLR2026
Alex on why AI + fully autonomous labs will compress drug discovery from five years to two:
"If scientists are armed with agentic infrastructure that has an intelligence that's a multiple of Einstein, they will do science faster, with higher probability, for lower cost, and have greater impact."
"AI coming into the discovery of medicines is one of the biggest unlocks we will see in history because it means we're no longer constrained by our hands or minds. We can identify all n number of hypotheses that could exist, figure out the best experiment to run, and run those 24/7."
"The sheer ability to be more comprehensive in our hypothesis set, and to run experimentation at unprecedented scale, means we're going to speed breakthroughs from timelines that take three to five years to timelines that take less than a couple of years."
Alex on why AI drug discovery companies need to generate novel data to succeed:
"AI models based on the research that's available is a lot of garbage in and garbage out."
"A lot of the recorded literature is actually incorrect. There's been tons of studies that show if you go try to replicate the experiments that are in the literature, you don't even get the same results."
"The AI companies that I believe are gonna be most set up for success are the companies with a novel way to generate science tokens that don't exist in the public domain."
On the train back from Fenway, I saw a dude get off in Newton wearing his marathon medal.
Running the Boston Marathon and taking the T home is THE MOST Boston thing ever and I love it so much.