Jacob Stern (@jacobastern) header
Jacob Stern (@jacobastern) profile picture

AI scientist @abscibio building virtual tissues previously: de novo antibody design (ORIGIN)

Vancouver, WA
Joined December 2017
Prediction: we will see a major Chinese model release in the next two days, timed to undercut the IPO Corollary: If no Chinese model drop, bullish update for frontier labs
74
My local AI agent just rescued my computer which was stranded without internet by an unlinked DNS resolution file. First tangible use case where local LLM beat Claude Code. Ready for the EMP.
3
1
4
222
Jacob Stern retweeted
I'm excited to announce CASMI 2026, the blind challenge hosted by @enveda on Kaggle to address one of the grandest challenges in life science: can you tell the structure of a small molecule from its mass spectrum? kaggle.com/competitions/enve…
9
1
14
1,229
Finally a more intuitive way to learn pLDDT/pAE? 😎 sokrypton.github.io/protein_… (Character idea from @HannesStaerk & Alex Waldherr)
56
287
44
1,402
166,313
It is an exciting time at Absci, with a pending Phase 1/2a readout (ABS-201 x hair loss) and rapidly improving, vertically integrated platforms for target discovery and antibody design.
2
193
The target discovery position is quite broad, spanning from genetics-based target identification to systems biology/perturbation modeling.
1
195
We have opened hiring for two AI Scientist roles @abscibio. The "AI Scientist - Antibody Design" position is for development of ORIGIN, our de novo antibody design platform. The "AI Scientist - Target Discovery" position is for development of ATLAS, our target discovery engine.
1
6
580
Claude and codex have gotten to the point where they invent their own project specific vocab. I use the Simplified Technical English prompt often when it gets out of hand. That seems to reasonate, the community put together skills and guides based on that idea Claude skill: github.com/danyuchn/asd-ste1… STE skill kit: github.com/woosal1337/blog/t… Writing styles: github.com/cfcosta/writing-s… Runbook skill: github.com/sdsheeks/ste100-s… STE checker: github.com/sourdough-bread/a… Enjoy! Happy slop combat
30
48
9
798
89,098
Jacob Stern retweeted
Introducing PerturbAI. Today we announced our emergence from stealth with the release of the world’s largest in vivo CRISPR data engine, interrogating the effects of thousands of genetic perturbations across 8 million cells throughout the whole brain. This dataset represents a new category of biological data: organism-level, circuit-resolved causal genomics leading to novel targets and therapeutics. By combining scalable in vivo CRISPR perturbation with AI, we model biological systems at unprecedented resolution and simulate therapeutic interventions before committing to expensive downstream development. We’re grateful to our collaborators at @NVIDIAHealth and @10xGenomics for helping make this landmark dataset possible. Read More: perturb.ai/news #CRISPR #AI #DrugDiscovery #FunctionalGenomics #Biotech
8
78
6
360
62,639
New SOTA on antibody-antigen complex prediction 🔥
🚀 Introducing Protenix-v1, the first open-source model achieving AF3-level performance Highlights: 🔹 Verified inference-time scaling behavior 🔹 RNA MSA & protein template support 🔹 Additional release: Protenix-v1-20250630 trained on a larger dataset 🔹 PXMeter v1.0.0 for transparent evaluation (6k+ complexes, time-split & domain-specific subsets) 🔗 Code: github.com/bytedance/Proteni… 🔗 Eval toolkit: github.com/bytedance/PXMeter 🔗 Online server: protenix-server.com
4
374
We are hiring AI scientists with a strong DL background, with experience training diffusion/flow matching models for molecular prediction. If you are interested in working on an AI team deeply integrated with a wet lab, please apply! DMs open, listing in the next post 👇
We @abscibio are excited to share Origin-1, our platform for de novo antibody design! We select "zero-prior" epitopes without guidance from or the availability of solved protein-protein complex structures and, using fewer than 100 designs per target, achieve success against four targets: COL6A3, AZGP1, CHI3L2, and IL36RA. We validate binding, developability, and function via multiple orthogonal assays and confirm atomic accuracy with Cryo-EM for two targets. We share detailed methods, including on the models underlying Origin-1, and release our data as a resource for the community. 📄 Read the preprint: absci.com/wp-content/uploads… 📊 Access our data: github.com/AbSciBio/origin-1
2
4
270