Artem Kushner

@GabriellaG439 gotta say thanks for introducing me to the Dijkstra archive, been reading the EWDs on and off toda.

New blog post: "A sufficiently detailed spec is code" I wrote this because I was tired of people claiming that the future of agentic coding is thoughtful specification work. As I show in the post, the reality devolves into slop pseudocode haskellforall.com/2026/03/a-suff…

this pairs well with that recent epiplexity paper

📢📢 Proteina-Complexa 📢📢 Atomistic Binder Design with Generative Pretraining and Test-Time Compute + Experimental Validation at Scale ⭐️ Project page (research.nvidia.com/labs/genair/pr…) for: 📜 Method paper (ICLR 2026 Oral) 🧬 Wet lab paper 🛠️ Code & models 📁 Data 🧵 Thread (1/n)

Radial.org is live! A new organization at @AsteraInstitute bringing together structural biologists, engineers, and ML scientists to redesign how we do science. @statnews has the story: statnews.com/2026/03/11/rad…

If you are working with the ribosome and are interested in the shape of its exit tunnel (for simulation, npet dynamics, etc.) here is a tool to get an Angstrom-fine mesh of it in under a minute. github.com/bioshape-analy…

You like discrete diffusion, but it's too slow? 🥀 You like test-time inference, but it's for continuous methods? 😩 We fixed it. Introducing Categorical Flow Maps: continuously sample discrete data in a single step 🚀💫 How? 🧵⬇️ 💪 Co-led with @FEijkelboom, @daan_roos_

NEW BLOG! @RuxandraTeslo, @kroetscha, @vientsek, @NeuroStats, and I have started a joint blog on Clinical Trials Abundance! We'll aim to publish weekly thoughts, commentary and ideas on how to make clinical trials more efficient & abundant. Subscribe: clinicaltrialsabundance.blog

@AnthropicAI i realize all the big boys are using claude code, but it would be very nice if one could collapse long code snippets in the claude.ai app.

🚀New paper: Discrete Feynman-Kac Correctors! It’s an inference time method that modifies discrete diffusion models via annealing, products, & reward-tilting. W/o training, it generalizes past the Ising critical temp. and boosts coding performance! arxiv.org/abs/2601.10403 🧵👇

Have a look at our new structure of co translational folding in yeast. This is collaborative work initialized by the Rospert lab from the @UniFreiburg . Structural work has been done by the amazing Lorenz Grundmann.

@DdelAlamo glhf, see you in two coffees...?

Only have 8 days left on my Claude Max plan so it's time to make the LigandMPNN/OpenMM-based replacement for Rosetta Relax that, for some reason, still doesn't exist

For encoding, current data formats are designed for static models, limiting biological insights and AI potential. We're creating standardized, machine-readable, and human-interpretable models of dynamic structures. See more details: diffuse.science/posts/encoding/

@DdelAlamo This is true. `mmcif` is a dumpsterfire

My hot take is that CIF and PDB formats are totally ill-equipped for adequately storing metadata describing how structures get generated and processed, and this problem will get worse as AI agents imbued with structural bioinformatics skills pass these files back and forth

1/🧵 We are very excited to release our new paper! From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence arxiv.org/abs/2601.03220 with amazing team @ShikaiQiu @yidingjiang @Pavel_Izmailov @zicokolter @andrewgwils

They provide a neat server visualizing all sorts of useful data about non-canonical amino acids and how they occur in PDB 👌 🤗 Paper: biorxiv.org/content/10.648… Server: mondet.tuebingen.mpg.de Very nice to easily see, e.g., how much training data of non-canonicals we have for each type and to quickly browse them!

Can we apply gradient descent to discrete changes? In our new #SIGGRAPHAsia paper, we show that gradient descent can work on shape grammars, as in CAD and procedural modeling, but only if the grammars are designed correctly!

AlphaFold predictions are valuable hypotheses and accelerate but do not replace experimental structure determination | Nature Methods nature.com/articles/s4159…

Fast and interpretable quantification of biological shape heterogeneity via stratified Wasserstein kernel biorxiv.org/content/10.110… #biorxiv_bioinfo

Don't worry if you missed MLSB! We recorded all of the talks!! 🕺💃 YouTube: @MLSBWorkshop/videos" target="_blank" rel="nofollow noopener">youtube.com/@MLSBWorkshop/…