iiinova — where your digital self keeps evolving

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

Oct 21–22, 2026. The event is free, hybrid, and international. A chance to make a mark. Direct Registration: luma.com/ku88xh92 Everything else: llmhackathon.github.io Come build the tools the next decade of science runs on and meet your new collaborators.

luma.com

4th Annual Open Scientific Intelligence Hackathon · Luma

Join us for an extraordinary convergence of minds where artificial intelligence meets scientific discovery. Building on our previous hackathons, we're…

010

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

Are you interested in running a site? 16 sites around the world hosted in 2025. A safe location, a room, wifi and maybe coffee. we handle the global program, judging, and the write-up. The site registration will open soon, but feel free to give us a heads up in Slack.

110

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

After 3 years, we know this creates brilliant outcomes, helps upskill those less familiar to the field, and just works. The event is open to grad students, postdocs, PIs, national lab staff, industry scientists, software developers, and more.

110

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

The event spans two days and is hybrid. You can join from a physical site or participate fully virtual. Teams form before the event and the day of, across time zones and build an amazing and tangible demo to be posted to social media

100

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

So, we wanted to give the event room to grow to match the frontier now and in the future. We also broadened the scope to more physical sciences to bring our community building model and what we've learned to other critical domains.

100

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

Why the rename? In 2023, we wondered "what can LLMs do to improve research in materials and chemistry?" Many showed the path forward using these new tools. But, after our last event, it's clear the frontier is more complex, relying heavily on agents, models, datasets, benchmarks, software, etc.

100

Ben Blaiszik @benblaiszik.bsky.social · 18/08/2026

The LLM Hackathon for Applications in Materials and Chemistry is now the Open Scientific Intelligence Hackathon, and our next event will be Oct 21–22, 2026! 🗓️ Oct 21–22 📍 Hybrid, online, in-person sites TBA As always, the event is free, hybrid, and worldwide. Register here llmhackathon.github.io

121

Ben Blaiszik @benblaiszik.bsky.social · 13/04/2026

Remember back when the NSF Graduate Fellowships were announced and there would be a solid week of the students posting about their great work? 🥲 Congrats to all the new fellows! www.nsf.gov/news/nsf-ann...

nsf.gov

NSF announces 2026 Graduate Research Fellowship Program award offers

The U.S. National Science Foundation (NSF) today announced that it has made offers and will award 2,500 Graduate Research Fellowships for the 2026-2027 academic year...

020

Ben Blaiszik @benblaiszik.bsky.social · 18/03/2026

Thanks for early feedback and support from: Will Engler, Owen Price Skelly, Ian Foster, Andrew Rosen, Naisargi Goyal, Isabella Furrick, Sihoon Choi, Ryan Szukalo, Sam Blau, Eliu Huerta, Shantenu Jha. See the code here and please reach out! github.com/Garden-AI/ro...

github.com

GitHub - Garden-AI/rootstock

Contribute to Garden-AI/rootstock development by creating an account on GitHub.

010

Ben Blaiszik @benblaiszik.bsky.social · 18/03/2026

Rootstock is early-stage but already in use on two clusters: Sophia at ALCF and Della at Princeton. If you have a project in mind that would benefit from easier use of multiple MLIPs, e.g., a benchmarking study or an MD campaign using different MLIP architectures, we could be an especially good fit.

100

Ben Blaiszik @benblaiszik.bsky.social · 18/03/2026

Do you use machine-learned interatomic potentials (MLIPs) on HPC clusters? Do you have a graveyard of condo environments to manage their incompatible dependencies? If so, we should talk. We built Rootstock to make it trivial to swap MLIPs when running atomistic simulation jobs with ASE/LAMMPS.

130

Reposted by Ben Blaiszik

Geoff Hutchison @geoffhutchison.net · 14/03/2026

There are already great translations into Hungarian, Indonesian, Romanian, Tamil, Ukrainian ... We could use more help with Chinese, German, Hindi, Japanese & Portuguese #chemsky

2168

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

Thank you Geoff! We should touch base sometime. I wonder if there are connections to be made between our model publishing efforts, TorchSim GPU MLIP software, and unique datasets (e.g., OMol25 we have millions of wave functions on hand), and more. Cheers!

010

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

Almost certainly every paper will be touched by AI/ML tools and the associated software stack in the not too distant future. That's just the way these technologies diffuse.

000

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

If you’d like to reuse or remix the figures for talks, papers, or slides, everything is available here: github.com/blaiszik/ml_...

github.com

GitHub - blaiszik/ml_publication_charts

Contribute to blaiszik/ml_publication_charts development by creating an account on GitHub.

130

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

To make the data easier to reuse, the repository includes: + Raw data queries, and data in csv format + Notebooks used to generate the plots + Both normalized and raw publication charts

100

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

This is quite a long-term trend as well. Across these fields, 5-year compound growth rates remain extremely high (Materials: 37.3%, Chemistry: 28.3%, Physics 27.9%) suggesting that the integration of machine learning into core scientific workflows is still accelerating.

100

Ben Blaiszik @benblaiszik.bsky.social · 10/03/2026

Updated AI/ML publication charts to include data from 2025! After what looked like an approaching S-curve top in 2024, the 2025 growth looks more bullish. In 2025, Web of Science data shows: 🔸 Materials: 12,987 papers +36.9% YOY 🔹 Chemistry: 16,522 papers +40.8% 🔸 Physics: 11,955 papers +27.5%

340

Ben Blaiszik @benblaiszik.bsky.social · 05/03/2026

Over the next month, we'll be releasing major updates to the Materials Data Facility! We will maintain support for all previous features, but there will be a lot of new ones. What are critical aspects you would like to see in a modern data repository? If you're interested in testing, send me a DM!

010

Ben Blaiszik @benblaiszik.bsky.social · 26/02/2026

Cloudflare rebuilt Next.js in a week with AI for $1,100. I wrote about how this could become a defining moment for the scientific community to erase decades of tech debt and build the software tools that we have imagined, to accelerate progress. Let's go! www.linkedin.com/pulse/what-w...

linkedin.com

What if We Just Built It All? A Call to Action to Reimagine Scientific Software in the Age of LLMs

Scientific software is the invisible infrastructure of modern research, and it's been starved of software engineering resources for decades. With LLMs and coding agents, that's changing.

030

Ben Blaiszik @benblaiszik.bsky.social · 12/02/2026

I was looking at my dissertation defense slides recently and was fondly like... "Wow, this is what it's like to focus on one project a year."

120

Ben Blaiszik @benblaiszik.bsky.social · 12/02/2026

The second panel will feature researchers from industry including Santiago Miret (Lila), Abhijeet Gangan (Periodic), Lily Kim (Microsoft Discovery), and Tim Erdmann (IBM) More details on the panel here, with registration: www.marda-alliance.org/2026-marda-a...

marda-alliance.org

2026 MaRDA Annual Meeting – Marda

010

Ben Blaiszik @benblaiszik.bsky.social · 12/02/2026

The Materials Research Data Alliance meeting is next week! Great panelists and opportunity for connection. Registration is free, and content is available online worldwide. Sign up to hear from key speakers: James Warren (NIST), John Schlueter (NSF), Andrew Schwartz (DOE), and Sean Donegan (AFRL)

110

Ben Blaiszik @benblaiszik.bsky.social · 03/02/2026

If your team works on ML for materials/molecular discovery and needs someone who has built physics-informed networks for real problems, check out Zhihao Feng. Genuine problem-solver, strong collaborator, adaptable. Here's an example of his work. Link: www.linkedin.com/posts/fengzh...

140

Ben Blaiszik @benblaiszik.bsky.social · 22/01/2026

What are the biggest software-solvable problems holding you back in your research? e.g., - There's no Python package to convert formats between X and Y - Running tasks on HPC systems requires expertise I don't have - Formatting or tracking the references for my paper is annoying

210

Ben Blaiszik @benblaiszik.bsky.social · 06/01/2026

As a kid, I was outdoors and in the woods all the time. The pandemic reminded me how awesome outside is. I still do as many meetings as possible on walks.

030

Ben Blaiszik @benblaiszik.bsky.social · 05/01/2026

or you'll just publish...much more. 😅 "If I had a million dollars We wouldn't have to eat Kraft Dinner But we would eat Kraft Dinner Of course we would, we'd just eat more"

110

Ben Blaiszik @benblaiszik.bsky.social · 05/01/2026

The 29th is a typo on the final day, it is in fact the 19th :)

000

Ben Blaiszik @benblaiszik.bsky.social · 05/01/2026

There will also be opportunities to connect with others interested in these topics through poster sessions and a community Slack. The event is online, so available around the world. Full speaker list soon!

100

Ben Blaiszik @benblaiszik.bsky.social · 05/01/2026

The 4th Materials Research Data Alliance meeting is set for Feb 17-19 on topics in materials + data + AI. This is a great opportunity to hear from speakers from academia, national labs, and industry. Register now (free - online): marda-alliance.org/2026-marda-a...

110

Ben Blaiszik @benblaiszik.bsky.social · 12/12/2025

I'm excited to help improve our understanding of catalysis and fusion materials via the Integrated Scientific Agentic AI for Catalysis (ISAAC) and CascAIde projects; and to see Rick Stevens and @ianfoster leading ModCon to develop/collect the models/data needed to advance science. Stay tuned!

000

Ben Blaiszik @benblaiszik.bsky.social · 12/12/2025

DOE announces $320M in new investments in the AI4Science! You can read more about the Genesis Mission, the American Science Cloud, the Transformational AI Models Consortium (ModCon), along with 51 AI in science application projects here and in the attached image. 🔗 energy.gov/articles/ene...

120

Ben Blaiszik @benblaiszik.bsky.social · 08/12/2025

Amazing week of results from AI4Science with NeurIPS and MRS. But, for those who couldn't make it, post a link to your research here we can boost you too!

042

Ben Blaiszik @benblaiszik.bsky.social · 30/11/2025

I saved this image from Stable Diffusion in Dec 2022 because it was amazing a model could output such detail. 😅

060

Ben Blaiszik @benblaiszik.bsky.social · 24/11/2025

Thanks to the 10 contributors! Thomas Loux, Curtis Chong, Rhys Goodall, Orion Archer Cohen, Will Engler, Abhijeet Gangan, Andrew_S_Rosen and others Bonus, a fun NanoBanana interpretation of the TorchSim paper

100

Ben Blaiszik @benblaiszik.bsky.social · 24/11/2025

New features include mixed periodic boundary condition (PBC) support, NVT Nose Hoover method, and more. This release also enables important integrations with QUACC and Atomate2 to proceed. More on that soon!

100

Ben Blaiszik @benblaiszik.bsky.social · 24/11/2025

TorchSim continues its growth aiming to be the high-performance engine for MLIP-powered atomistic simulation. With this release (4.1), there improvements across the entire stack with new features, bug fixes, and improved documentation.

162

Ben Blaiszik @benblaiszik.bsky.social · 21/11/2025

So your data are available upon reasonable request? Well, we are making some reasonable requests - at scale. :) 1. Search literature (currently stubbed) 2. Enumerate papers, extract contacts 3. Send email w/ data drop location 4. Parse data Does anyone want to help productionize this?

051

Ben Blaiszik @benblaiszik.bsky.social · 19/11/2025

Check out our updated website, publish your data, or leverage the 600 TB of data we have on hand at www.materialsdatafacility.org

000

Ben Blaiszik @benblaiszik.bsky.social · 19/11/2025

- First, a new section showing how to use MDF's Foundry datasets. Load ML-ready datasets with just a few lines of code - Second, a clean hero, with updated Navbar and clear Calls-to-Action - Third, a more compact and intuitive view of Foundry dataset contents, browsable from the search interface

100

Ben Blaiszik @benblaiszik.bsky.social · 19/11/2025

Within Cursor, Gemini 3, and Claude missed simple things that caused component errors. Sonnet fixed these immediately in CC, showing the importance of the harness. There were lots of little updates, but the attached images show a few specifically

100

Ben Blaiszik @benblaiszik.bsky.social · 19/11/2025

2. The model harness matters...a LOT. That means that you might see vastly different performance e.g., in Cursor vs Antigravity (Google's IDE) vs Droid. 3. The cost was low. This update took 2h, $0.90 Cursor credits, 2 free Antigravity prompts and 1 Claude Code (CC) prompt.

100

Ben Blaiszik @benblaiszik.bsky.social · 19/11/2025

Yesterday I did a glow-up of the entire Materials Data Facility front end using Gemini 3 in less than 2 hours. A few things I learned followed by few snapshots of the updates. What did I learn? 1. Gemini 3 feels clearly ahead in terms of front-end design. Less AI slop, and more focus.

120

Ben Blaiszik @benblaiszik.bsky.social · 12/11/2025

Northern Lights over the western Chicago suburbs at 9:15. Clearly visible by eye, lasted about 10 mins. 🤩

0112

Ben Blaiszik @benblaiszik.bsky.social · 01/11/2025

So glad you are getting some relief from all of this! 💜

110

Ben Blaiszik @benblaiszik.bsky.social · 01/11/2025

Did a quick experiment last night to see how far kids run on Halloween by giving my son a phone. The answer was 5.3 miles.

020