Super proud to see this finally out! Amazing work with @n_gao96 and my other lovely colleagues from @cusp_ai ! Looking forward to seeing what the community will actually build with this! ☕️☕️☕️🚀
Molecular simulation is the backbone of drug discovery and materials science, but it’s notoriously complex.
Today, we’re excited to open source kUPS - a molecular simulation engine built for the AI era, optimized for GPU in collaboration with @nvidia.
The dogma of the last decade in computer vision is that all classical algorithms should be replaced by neural networks. This has been very productive but, counterintuitively, with the latest frontier models, I find myself sometimes turning back to classical methods.
If you have the training data, the dogma still holds. But there are situations where data is hard to get, there exist good classical algorithms for solving a problem, optimization methods are just better, or where some simple code just does the thing you want.
When writing code was time consuming and error prone, training everything made sense. Now, that a frontier model can code up and test solutions quickly, getting a solution through coding can be faster than gathering training data and training a model. Consequently, I think there will be more hybrid approaches that blend trained models and code.
We now have generic, high-level, planners that can execute sequences of lower-level tasks to solve large problems. These lower-level tasks may be solved by a neural network, a classical method, or specialized code written on the fly.
It never pays to be religious about technology. The future is probably hybrid. Pick the thing that works for you.
Proud to partner with MARVEL-AI, Singapore’s new S$150M platform for AI-driven material discovery, led by @ASTARsg and announced by Deputy Prime Minister @GANKIMYONG 🇸🇬
Our AI platform + their autonomous labs = new materials designed for impact at scale
Our co-Founder & CTO @wellingmax was there for the launch. 📷
📣 Read more: theedgesingapore.com/digital…
Like many people, I am really fascinated by turbulence.
Numerical choices heavily impact the the characteristics of turbulence, a lot.
For us as an AI4Science community, this matters because we use simulated flows to train and evaluate our models.
A thread on turbulence 🧵
We're hiring Application Scientists 🧪
At CuspAI, we're using AI to discover the materials the world needs, for energy, clean water, the future of compute and carbon capture.
Our Scientific Discovery teams are growing, and we're looking for Application Scientists to join us in Electronics, Energy and Environment.
You'll own the science on partner projects from scoping to results, working with ML researchers, platform engineers and lab teams, and translating between experiment, simulation and machine learning.
We're especially interested in hands-on experience in:
🔷 Semiconductors, thin films and materials for electronics
🔷 Catalysis, batteries and photovoltaics
🔷 Porous materials, formulations, water treatment and separation
🔷 Lab automation, cheminformatics and optimisation
Industry experience is a big plus. You don't need to cover a whole pillar: real depth in one area is what counts.
🌍 Europe (Cambridge, London, Amsterdam or Berlin)
👉 jobs.ashbyhq.com/cuspai/a755…
🇸🇬 Singapore
👉 jobs.ashbyhq.com/cuspai/10d8…
Know someone who'd be a great fit? Tag them in the comments or share this post 🙏
Public preview of BioEmu2! @MSFTResearch AI for Science’s new foundation model for biomolecular structure, dynamics and function. Open weights, MIT license.
github.com/microsoft/bioemu2
Movie: BioEmu2-simulated binding of intrinsically disordered peptide PUMA to MCL-1 receptor. 🧵⬇️
Where computational irreducibility makes particle kinetics infeasible, it gifts us fluid mechanics to compensate.
Where it makes neurophysiology infeasible, it gifts us psychology to compensate.
Mechanistic interpretability is infeasible. What have we been gifted to compensate?
Both are true at the same time:
Entire disciplines will be disrupted by labs with large resources.
Scientific discovery is democratised as asking the right questions and spending a few dollars worth of tokens can replace specialist skills
Either way things are speeding up
Mathematician Terence Tao:
"we're seeing the end of the romantic era of mathematics, the age of heroes. we're moving into an industrial era where math becomes large-scale, done in big teams like the other sciences"
We're entering, rather unwillingly, a new era where exciting new things are possible, but some things are being lost
Die Todesstrafe ist unzulässig, da sie die Unverletzlichkeit und Würde des Menschen verletzt. Das Gemeinwohl kann geschützt und den Anforderungen der Gerechtigkeit kann Rechnung getragen werden, ohne auf die Todesstrafe zurückzugreifen: Es wurden wirksame Haftsysteme entwickelt, die die Bürger schützen, den Schuldigen aber gleichzeitig die Möglichkeit zur Resozialisierung nicht vollständig nehmen. #NoDeathPenalty
Leaving a frontier lab is the biggest reality check of all time.
I spent my days at @OpenAI believing I was peering into the inevitable future.
A few months out, I think I had it completely backwards.
Wiederhole mich: Man kann die Linke politisch beknackt finden, aber für Demokratie & Verfassung ist sie keine Gefahr. Im europäischen Vergleich ist die deutsche Linke sogar moderat, das gilt auch für Haltung zu Israel.
AfD hingegen ist eine der rechtsextremsten Parteien Europas.
Most people without experience fail to appreciate just how hard (i.e., nearly impossible) it is to simulate the physical world with high fidelity.
It's exactly the same as people who've never been in a wet-lab assuming that you can just think your way to cooking up a new virus.
Interested in pursuing a PhD with me at @UvA_Amsterdam on mechanistic interpretability, AI for science, or AI safety? NWO Mozaiek is funding PhD students with non-Western migration backgrounds: nwo.nl/en/calls/mosaic-30