Srihari Ganesh

Member of Technical Staff @axiommathai. Prev: ML Scientist @nablabio, Harvard '24 (Stat+Biochem).

[Reticular (YC F24) @ReticularAI]

Genomes are long but redundant. Not great for pretraining.

To fix that, we've developed a lossless tokenization strategy that shortens sentence length by >2000x.

For pretraining, we achieve a 0.55 validation CE on All of Us.

Public replication on 1KG linked below.

[Axiom @axiommathai]

Most think vibecoding at scale is slop. Here’s a counterexample.

Today we release a 16M-line formalization of the entire Stacks Project, critical infra underlying modern math, with optimized code quality and its highly interconnected structure preserved.

A mathematical city 🌆

[Nabla Bio @nablabio]

Today we’re thrilled to announce JAM-2 — the first AI model capable of generating drug-quality antibodies straight from the computer, with industry-leading success rates.

> Drug-like affinities: Picomolar to single-digit nanomolar antibody binders for half of 26 targets while testing <45 designs each.

> Unlocking hard targets: Up to 11% success rate for direct on-cell GPCR binders; top antibody hits in the single-digit nanomolar range.

> Unprecedented epitope breadth: JAM-2 routinely designed antibodies that hit 30–70% of user-defined epitopes, now enabling intentional design of biology — not chance discovery.

> Drug-like developability: Over 50% of antibody designs passed core industry developability criteria with zero optimization.

> Massive leverage: A four-person team prosecuted 16 targets in parallel in < 1 month.

JAM-2 is the first de novo antibody design capability ready for front-line use in drug discovery, matching or surpassing traditional discovery approaches.

We’re already deploying JAM-2 with multiple large pharma partners and seeing excellent results.

If you’re interested in partnering on molecule development or accessing JAM-2, contact [email protected].

Read more in our whitepaper (link below)

[Surge Biswas @SurgeBiswas]

Test-time scaling has arrived for biological design. Excited to share new work from Nabla Bio.

By scaling test-time compute, we've achieved remarkable leaps in de novo antibody design against difficult drug targets.

[Surge Biswas @SurgeBiswas]

If you're at BioLogic in San Diego next week, check out Adithya's talk on our work in de novo antibody design!

[Harvard Statistics Department @HarvardStats]

The Group for Undergraduates in Statistics at Harvard (GUSH) and the Harvard College Women's Center hosted a Women in Statistics and Data Science Panel as part of Women’s Week 2023 and Women’s History Month. Read more about this event here: https://t.co/1fsfKtAziv

[Harvard Statistics Department @HarvardStats]

Thanks to the undergraduate organization GUSH and the Harvard Statistics Graduate Council, students, postdocs, faculty, and staff enjoyed a cookie social!

[Harvard Statistics Department @HarvardStats]

GUSH and @HCWC_WomensCtr put together a fantastic Women in Statistics &amp; Data Science Panel featuring statisticians/data scientists in academia, journalism, consulting &amp; investment mgmt. Thank you to the panelists Amanda Cox, Sowmya Rao, Kim Nguyen &amp; Kelly McConville! #IWD2023

[MIT Jameel Clinic for AI & Health @AIHealthMIT]

#TeamOctavian is ready to run the @CambClassic5Ks, rain 🌧️ or shine ☀️ 🏃‍♀️🏃‍♂️

[MCB_Harvard @MCB_Harvard]

Alexander Chen, Srihari Ganesh, Clarence Ndubisi, and Will Nickols Elected to Phi Beta Kappa (PBK) @PhiBetaKappa @srihariganesh @ccndubisi https://t.co/wzfAPswRYe