🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

With the start of my first batch as a visiting partner at Y Combinator, this felt like a good time to write my founder story about our journey at Reverie Labs. It's a bit of a long one. ankitg.me/blog/2025/01/0…

I'm spending this month in SF. Would be excited to connect in person. Ping me.

We have a position open for an ML Scientist role to take part in one of the most ambitious undertakings in AI-driven protein design. If you have worked on frontier models (in or outside biology), or protein design, join us and find out. dynotx.com/careers/?gh_ji…

So far, language models applied to biological sequences accelerate and democratize specialized (but doable) bioinformatics tasks. Often not at SOTA. They are still in their GPT-2 phase (Thought probably data/architecture limited not compute). Unlike Alphafold, which was a real

Don’t miss our co-founder Sam Sinai’s talk at #GTC2025! Sam will share Dyno’s latest progress on LEAP—our technology bridging foundation models and therapeutic impact—and demo DynoFold, a new AI model powered by NVIDIA AI that efficiently predicts protein dynamics directly from

Have we hit a "scaling wall" for protein language models? 🤔 Our latest ProteinGym v1.3 release suggests that for zero-shot fitness prediction, simply making pLMs bigger isn't better beyond 1-4B parameters. The winning strategy? Combining MSAs & structure in multimodal models!

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.

Was a huge pleasure to host Anthony Costa at a packed event with the Dyno Therapeutics, Inc.'s team at NY Tech Week!

Really impactful and exciting progress by folks at NVIDIA Healthcare

Protein structure ≠ protein folding/dynamics The new north star: energy‑landscape learning, rather than static structure prediction Next wave: AI that learns free energy for "true" inverse folding

Israel is just warning people to immediately evacuate a large patch of Tehran. Say the equivalent of Williamsburg. My septuagenarian mother lives in the equivalent of Brooklyn, 2 miles from this boundary. Many of my friends live inside that patch.

Tehran has upwards of 15M metro population. The roads out have been blocked by traffic all day. Where and how do you think these people will evacuate?! My 72 yo mother is literally there right now, doesn't even have a car. I personally know 100s of people in a similar situation.