An interactive dashboard displays a UMAP of the tested generative samples, and various properties of the discovered landscape.

JURA is launching a new target intelligence dashboard called MESA jurabio.com/mesa with a nice interactive case study from our recent TCRm campaigns. Go check it out -- we'll open to crowd-sourced targets soon. @jura.bsky.social

One billion simultaneous experiments — JURA Bio, Inc. For decades, drug discovery has been constrained by a simple fact: experiments are expensive, so you have to guess well. We built a system where you don't have to guess — testing a billion distinct m...

Prediction is overhead when verification is cheap.

We've spent years building a system where verification is cheap -- generationally so.

This changes the logic of discovery in ways that are easy to underestimate. I tried to write a little about what that means: www.jurabio.com/blog/onebill...

Accelerated Learning on Large Scale Screens using Generative Library Models Biological machine learning is often bottlenecked by a lack of scaled data. One promising route to relieving data bottlenecks is through high throughput screens, which can experimentally test the acti...

Paper: arxiv.org/abs/2510.16612
Blog: www.jura.bio/blog/leavs
Team: @lizbwood.bsky.social @highvariance.bsky.social @mgollub.bsky.social

Our LeaVS preprint is up! We demonstrate empirically & prove theoretically that LeaVS can dramatically accelerate learning, increasing the effective dataset size by orders of magnitude. Really useful biological assays have intractable bottlenecks: make the most of what you can get.
@jura.bsky.social

How do you train AI on protein data when functional sequences are astonishingly rare? Traditional iterative optimization fails because most measurements show zero.

Our newly announcing learning system, LeaVS, shows a way.

LeaVS: Accelerating learning for biological AI — JURA Bio, Inc. A fundamental lesson of modern AI is that scale is essential: training bigger models on bigger datasets unlocks new capabilities. A fundamental lesson of AI engineering is that scaling up isn't trivia...

You can read more in our post at www.jurabio.com/blog/leavs; preprint forthcoming.

@jura.bsky.social @eliweinstein.bsky.social @mgollub.bsky.social @highvariance.bsky.social

A fundamental lesson of modern AI is that scale is essential: training bigger models on bigger datasets unlocks new capabilities. A fundamental lesson of AI engineering is that scaling up isn't trivial: it is not just a matter of spending more money and resources.

Our head of ML research Eli Weinstein @eliweinstein.bsky.social is recruiting the first PhD student for his new group at @dtu.dk in Denmark.

Help create the cutting edge of machine learning methods for molecular sciences.

Full job posting in link.

So excited about this - we did iterative generative design at large scale with variational synthesis, and got human scFv candidates against some of the hardest therapeutic targets around.

Hardware/wetware codesigned data loop VISTA makes use of generative model sampling and synthesis "on chip" on-board by leveraging oligosynthesis setup shown here.

The biggest challenge for AI in biology isn't just models, it's the data used to train them. Standard biological data isn't built for AI. To unlock generative AI for drug discovery, we must rethink how we generate and capture data. 1/

This week is the BIO conference in Boston's Seaport. If you are in town for it and are interested in meeting to learn more about @jura.bsky.social please feel free to send a DM! Would love to see you.

I think we’ve now manufactured more AI drug designs than OpenAI has generated responses, but plenty of bottlenecks remain to be innovated upon ‘til delivering the safest and most effective therapeutic is as easy as clicking print.

DM if you’re interested in joining us at @jura.bsky.social.

Congratulations to our Director of Machine Learning Research, who will be starting is group at @dtu.dk chemistry this fall.

If there's such a thing as AI for drug design & development on easy mode, then it's probably in being able to build and test every design.

Check out our journey to the 10 quadrillion AI-designed antibody library now; we're going to be sharing some big updates soon:

▶️ www.jura.bio/blog/variati...

Our Head of ML Research @eweinstein.bsky.social will be presenting tomorrow at CMU if you're local; the talk is also available on zoom if you follow the link.

At UPenn CIS today, JURA's Head of ML Research will be presenting variational synthesis (among other methodological advances) @eweinstein.bsky.social

Excited to be one of the keynote speakers at this weekend's Future of Care 2025 conference, hosted at The Rockefeller University in NYC. If you're in town, it would be wonderful to see you there! Details in comments.

This Thursday, @eweinstein.bsky.social is traveling to the University of Chicago to give a joint Statistics & Data Science colloquium. If you've been interested in learning more about #variationalsynthesis and are local, I encourage you to attend!

Event link: stat.uchicago.edu/events/event...

Excited to see a lot of our team joining 🦋! If you're interested in following them, you can find them at:

@lizbwood.bsky.social
@eweinstein.bsky.social
@mgollub.bsky.social
@highvariance.bsky.social

Plus our scientific advisor:

@lestermackey.bsky.social

Hopefully many more to come!

Looking forward to this event tomorrow. It's been so wonderful visiting Basel so far: it probably won't come as a surprise, but the city is really gorgeous and with wonderful trams.

Here are two pictures from the university's botanical gardens as the sun was setting.

New @jura.bsky.social merch dropped today! Dm to reserve yours. Very generous one-size-fits-all.

Tomorrow in Baltimore, head to the Johns Hopkins University Computer Science Department Seminar Series for more on #VariationalSynthesis with
@eweinstein.bsky.social.

Full details: www.cs.jhu.edu/event/cs-sem...

Introducing: Manufacturing aware generative model architectures — JURA Bio, Inc. Today, we introduce Variational Synthesis , the first manufacturing-aware generative AI model architecture for biological sequence models. Generative variational synthesis models can be used to build...

I'm really happy to re-introduce myself here 🦋. At @jura.bsky.social, we are building an AI-native biotech so that we can more quickly create safer, more effective genomic medicines for more and more patients.

Recently, we announced:
www.jurabio.com/blog/variati...

And:
www.jurabio.com/blog/caire

Today (Nov 12) is the last day to apply to the #TermeerFellows program. If you are a first-time founder building a patient-focussed company, I couldn't recommend the program, or its surrounding community, more highly. Here's the link:

termeerfoundation.slideroom.com#/login/progr...

Qs? DM me.

Julie came from Boston to pose for @jura.bsky.social headshots this weekend. Hopefully we’ll have some nice news coming out this week or next! In the meantime, here’s one of the pictures that didn’t make it…

Real-life picture from our new startup space in Basel! And every time I've met with a local representative, I've left with a box of chocolates. #DreamsDoComeTrue