Jeff Ruffolo

AI has two modes in drug discovery. Accelerate: moving faster through the existing playbook. Unlock: opening frontiers that weren't possible before. Excited to announce Profluent is partnering with Eli Lilly, the global pharma powerhouse, to unlock breakthrough medicines for patients. It's a big deal beyond the numbers ($2.25B + royalties): we’ll get to use our frontier AI models and foundational datasets to design proteins focused on large gene insertion, a therapeutic moonshot. Proteins govern almost everything in biology. We've built a generalizable AI platform to design all proteins. Onward!

Today, @ProfluentBio and @EliLillyandCo announce a multi-$B partnership to use our AI models to design custom recombinases, a new class of gene editor capable of large-scale DNA editing across multiple diseases. I'm so proud of the team - Profluent is an n=1 company.

With Lilly, we're applying that to recombinases, proteins capable of inserting kilobases of DNA into a genome and designed to target specific genetic disorders. Congrats to the whole team on this milestone!

This project is a clear expression of what generative protein models are best at. Train on nature's full catalog of sequences, learn the patterns underneath,  then design new proteins for problems nature didn't solve.

Excited to share that we’ll be working with @EliLillyandCo to develop AI-designed recombinases for genetic medicine.

Looking for strong engineering, hands-on PLM experience, and real protein domain knowledge. If this sounds like your background, please reach out or apply directly!

We're a small, interdisciplinary team — ML scientists, protein design scientists, biologists, bioinformaticians — and this person would work across all of those groups. High-ownership IC role where what you build goes directly into design campaigns that hit the lab.

We're hiring on my team at @ProfluentBio: Protein Design Scientist, Agentic Workflows. Protein language models learn the evolutionary rules of natural sequences, but turning that into a design pipeline that works end-to-end is a different problem. job-boards.greenhouse.io/profluent/jobs…

Today we’re announcing $106M in new funding led by Altimeter Capital and Bezos Expeditions. This brings our total to $150M to scale our frontier AI models which make biology programmable. Our frontier models have generated functional proteins (Nature Biotech, 2023), created the first CRISPR system designed from scratch (Nature, 2025), and showed clear scaling behavior (NeurIPS spotlight, 2025). The opportunities ahead are unimaginable. If you’re excited by shaping the future of biology – join us in pushing the science forward. Forbes: forbes.com/sites/amyfeldm… Press Release: businesswire.com/news/home/2025… -- Nature Biotech, 2023: nature.com/articles/s4158… NeurIPS spotlight, 2025: biorxiv.org/content/10.110… Nature, 2025: nature.com/articles/s4158…

AI can bring high quality genetic tools within reach for rare diseases. Today, Profluent is partnering with the Rett Syndrome Research Trust (RSRT) to design personalized genomic medicines for Rett syndrome. Our aim is to engineer compact editors that can fit into a single AAV delivery capsid, a key step toward reaching the central nervous system. Once there, our editors will need to precisely correct recurrent MECP2 hotspot mutations that drive Rett syndrome pathology. Rare disease used to mean rare attention. Specialized genetic tools were reserved only for the most common conditions. AI is changing that, bringing state-of-the-art biotechnology to a rare neurodevelopmental disorder that touches families every day. Profluent and RSRT are excited to bring the ProGen3 model and the full strength of our platform to bear on Rett syndrome. We’re working relentlessly toward a future when AI helps to make the best science accessible to all patients. 🔗 reverserett.org/news/articles/…

Protein language models just got an upgrade. Meet Profluent-E1: a free, flexible, frontier protein sequence encoder. E1 is built with retrieval augmentation to learn from multiple sequences. Models trained over 4T tokens with only 150M-600M params, E1 is SOTA for zero-shot functional and unsupervised structural tasks. It raises the bar for protein representation learning and is freely available today.

Learn more about E1 and start using the model: Paper: biorxiv.org/content/10.110… Blog: profluent.bio/showcase/e1 Code: github.com/Profluent-AI/e1 Congrats to @successar_nlp and the rest of the team @AadyotB @daggerlog @thisismadani on the release!

These evaluations are just the beginning. At @ProfluentBio, we’ve already found E1 to be a compelling alternative to existing encoders. Our hope in releasing these models is for others to boost their own workflows and provide feedback that helps us continue to push the frontier.

I’m excited to share Profluent-E1, our internal encoder model for protein sequences, with the protein-ML community! The E1 series of models can be used as a drop-in replacement to upgrade a broad range protein modeling and design of workflows. biorxiv.org/content/10.110…