Kyle Cranmer
@kylecranmer.bsky.social
Director Data Science Institute @UWMadison, Professor of Physics,
EiC @MLSTjournal. Physics, stats/ML/AI, open science.
EiC @MLSTjournal. Physics, stats/ML/AI, open science.
I migrated my somewhat janky / customized pelican3.7 website to pelican4.11 using Claude Code. Migrated to a clean repo, streamlined GitHub Pages deployment, new features!
theoryandpractice.org/2025/12/migr...
theoryandpractice.org/2025/12/migr...
Migrating with Claude
I migrated my personal website from Pelican 3.7 to Pelican 4.11 using Claude Code, Anthropic's agentic AI coding tool. Here's what changed and how it went.
theoryandpractice.org
December 27, 2025 at 6:27 PM
I migrated my somewhat janky / customized pelican3.7 website to pelican4.11 using Claude Code. Migrated to a clean repo, streamlined GitHub Pages deployment, new features!
theoryandpractice.org/2025/12/migr...
theoryandpractice.org/2025/12/migr...
Update... 3D travel "Worldline". Something I've wanted to visualize for years. It makes this automatically from a folder of images.
GitHub: github.com/cranmer/trav...
Gallery: theoryandpractice.org/travel-map/
GitHub: github.com/cranmer/trav...
Gallery: theoryandpractice.org/travel-map/
December 25, 2025 at 1:42 AM
Update... 3D travel "Worldline". Something I've wanted to visualize for years. It makes this automatically from a folder of images.
GitHub: github.com/cranmer/trav...
Gallery: theoryandpractice.org/travel-map/
GitHub: github.com/cranmer/trav...
Gallery: theoryandpractice.org/travel-map/
ok, for fun I asked Claude Code to make me a tool where I can drop in some photos from my travels and have it make me a travel map. Thinking about my year in review.
December 24, 2025 at 12:44 AM
ok, for fun I asked Claude Code to make me a tool where I can drop in some photos from my travels and have it make me a travel map. Thinking about my year in review.
Reposted by Kyle Cranmer
A bipartisan group in Congress formally pushed back against administration plans to dismantle NCAR, while the country experiences a record warm Christmas week with monthly records - and CA braces for more serious flooding and wind (and eventually snow). More: tinyurl.com/3b2tsp7m
Bipartisan Congressional coalition formally requests continued funding for National Center for Atmospheric Research
Next in series of storm systems into California will likely be most impactful with potential flooding and strong winds through Christmas Eve, while incredible late December warmth spreads east
tinyurl.com
December 23, 2025 at 6:23 PM
A bipartisan group in Congress formally pushed back against administration plans to dismantle NCAR, while the country experiences a record warm Christmas week with monthly records - and CA braces for more serious flooding and wind (and eventually snow). More: tinyurl.com/3b2tsp7m
Check it out! @UWMadison is ranked #5 in research expenditures. NSF's Higher Education Research and Development (HERD) rankings
ncses.nsf.gov/surveys/high...
ncses.nsf.gov/surveys/high...
December 23, 2025 at 7:24 PM
Check it out! @UWMadison is ranked #5 in research expenditures. NSF's Higher Education Research and Development (HERD) rankings
ncses.nsf.gov/surveys/high...
ncses.nsf.gov/surveys/high...
Over the last few days I was playing around with
claude code. I uploaded a photo of a board game my family plays and had it build an RL environment, a UI, and do some reproducible analysis of strategies with MyST markdown.
github.com/cranmer/cass...
theoryandpractice.org/cassville-ch...
claude code. I uploaded a photo of a board game my family plays and had it build an RL environment, a UI, and do some reproducible analysis of strategies with MyST markdown.
github.com/cranmer/cass...
theoryandpractice.org/cassville-ch...
December 23, 2025 at 5:31 PM
Over the last few days I was playing around with
claude code. I uploaded a photo of a board game my family plays and had it build an RL environment, a UI, and do some reproducible analysis of strategies with MyST markdown.
github.com/cranmer/cass...
theoryandpractice.org/cassville-ch...
claude code. I uploaded a photo of a board game my family plays and had it build an RL environment, a UI, and do some reproducible analysis of strategies with MyST markdown.
github.com/cranmer/cass...
theoryandpractice.org/cassville-ch...
Reposted by Kyle Cranmer
This is not just "NCAR boosterism," folks. There are multiple empirical measures by which this is correct--and also, if you took a random sampling of atmospheric & Earth scientists from around the world, I'd bet that a substantial majority of them would agree. NCAR is truly irreplaceable.
Bucalacchi stresses NCAR's reach is unique globally: "if you look at the totality of activities from weather, water, climate, space, weather, atmosphere, chemistry- no one single institution covers that breadth and depth of across observation, high performance computing and modelling and tradition."
December 18, 2025 at 11:04 PM
This is not just "NCAR boosterism," folks. There are multiple empirical measures by which this is correct--and also, if you took a random sampling of atmospheric & Earth scientists from around the world, I'd bet that a substantial majority of them would agree. NCAR is truly irreplaceable.
Reposted by Kyle Cranmer
The LHC community has had more than an exabyte of data and simulation globally distributed, but now @cern.bsky.social has hit an exabyte of experimental observations at its data centers. At these scales, analysis of data from the HL-LHC upgrade will require new approaches.
home.cern/news/news/co...
home.cern/news/news/co...
December 19, 2025 at 1:05 AM
The LHC community has had more than an exabyte of data and simulation globally distributed, but now @cern.bsky.social has hit an exabyte of experimental observations at its data centers. At these scales, analysis of data from the HL-LHC upgrade will require new approaches.
home.cern/news/news/co...
home.cern/news/news/co...
Reposted by Kyle Cranmer
It is hard to overstate how critical @ncar-ucar.bsky.social is to climate science in the US and around the world. It's the beating heart of our field. Generations of scientists have trained there, and almost everyone I know relies on deep collaborations with NCAR scientists. It's end is unthinkable.
Exclusive: The Trump administration is moving to dismantle the National Center for Atmospheric Research in Colorado, according to a senior White House official, taking aim at one of the world's leading climate research labs.
Trump moves to dismantle major US climate research center in Colorado
The Trump administration is breaking up the National Center for Atmospheric Research, taking aim at one of the world's leading climate research labs.
bit.ly
December 17, 2025 at 2:50 AM
It is hard to overstate how critical @ncar-ucar.bsky.social is to climate science in the US and around the world. It's the beating heart of our field. Generations of scientists have trained there, and almost everyone I know relies on deep collaborations with NCAR scientists. It's end is unthinkable.
Reposted by Kyle Cranmer
These people are nuts. NCAR developed the GPS dropsonde, which revolutionized the understanding of TC structure, improved forecasts, and validated remote sensing platforms, as documented in over 400 peer-reviewed publications in the last 25 years. And that’s just one of NCAR’s countless advances.
Exclusive: The Trump administration is moving to dismantle the National Center for Atmospheric Research in Colorado, according to a senior White House official, taking aim at one of the world's leading climate research labs.
Trump moves to dismantle major US climate research center in Colorado
The Trump administration is breaking up the National Center for Atmospheric Research, taking aim at one of the world's leading climate research labs.
bit.ly
December 17, 2025 at 2:09 AM
These people are nuts. NCAR developed the GPS dropsonde, which revolutionized the understanding of TC structure, improved forecasts, and validated remote sensing platforms, as documented in over 400 peer-reviewed publications in the last 25 years. And that’s just one of NCAR’s countless advances.
Reposted by Kyle Cranmer
Our fantastic #MachineLearningScienceandTechnology editor-in-chief @kylecranmer.bsky.social giving a shout out to our collaboration with the #ML4PS workshop at #NeurIPS2025 @iopp-mlresearch.bsky.social Thank you, Kyle! 🥳 #machinelearning #chemistry #physics
December 7, 2025 at 12:09 AM
Our fantastic #MachineLearningScienceandTechnology editor-in-chief @kylecranmer.bsky.social giving a shout out to our collaboration with the #ML4PS workshop at #NeurIPS2025 @iopp-mlresearch.bsky.social Thank you, Kyle! 🥳 #machinelearning #chemistry #physics
Reposted by Kyle Cranmer
If you are interested in diffusion models for dynamical systems and posterior inference, stop by our poster at ML4PS workshop (@neuripsconf.bsky.social) at 11am!
Come and see how we apply modern methods to the challenging field of global weather inference!
Come and see how we apply modern methods to the challenging field of global weather inference!
December 6, 2025 at 6:53 PM
If you are interested in diffusion models for dynamical systems and posterior inference, stop by our poster at ML4PS workshop (@neuripsconf.bsky.social) at 11am!
Come and see how we apply modern methods to the challenging field of global weather inference!
Come and see how we apply modern methods to the challenging field of global weather inference!
An unexpected shoutout from @kyunghyuncho.bsky.social in his NeurIPS keynote. @glouppe.bsky.social & I collaborated with him to bring ideas from natural language to high-energy physics back in 2016.
arxiv.org/abs/1702.00748
arxiv.org/abs/1702.00748
December 5, 2025 at 4:50 PM
An unexpected shoutout from @kyunghyuncho.bsky.social in his NeurIPS keynote. @glouppe.bsky.social & I collaborated with him to bring ideas from natural language to high-energy physics back in 2016.
arxiv.org/abs/1702.00748
arxiv.org/abs/1702.00748
Reposted by Kyle Cranmer
Discover the '2025 Early Career Researcher Best Paper Prize' and the '2025 Best Paper Prize' winning articles and make a nomination for the 2026 Best Paper Prizes here: https://ow.ly/f10S50XALIq
December 4, 2025 at 12:00 PM
Discover the '2025 Early Career Researcher Best Paper Prize' and the '2025 Best Paper Prize' winning articles and make a nomination for the 2026 Best Paper Prizes here: https://ow.ly/f10S50XALIq
Reposted by Kyle Cranmer
Beautiful keynote by @gaelvaroquaux.bsky.social at @euripsconf.bsky.social .
December 5, 2025 at 12:50 PM
Beautiful keynote by @gaelvaroquaux.bsky.social at @euripsconf.bsky.social .
Frustrated by trying to publish in the large conferences? MLST is a great venue to share research at the intersection of AI/ML and the physical sciences.
🌟Discover the latest research that has got the scientific community talking🌟
Be one of the first to unlock MLST's brand new research hub - where we will share the latest and most cited articles published in the journal, all in one place.
🔓Unlock the research: https://ow.ly/7yM550XtYnq
Be one of the first to unlock MLST's brand new research hub - where we will share the latest and most cited articles published in the journal, all in one place.
🔓Unlock the research: https://ow.ly/7yM550XtYnq
December 5, 2025 at 1:08 PM
Frustrated by trying to publish in the large conferences? MLST is a great venue to share research at the intersection of AI/ML and the physical sciences.
Reposted by Kyle Cranmer
I cannot agree with this more, and I'd add that I've never been mad that I spent a lot of time learning something thoroughly — and frequently it does take me a lot of time — whereas I've often been mad (and embarrassed) that I didn't take the time to learn something that I should know.
If you feel like you are behind or underprepared, the only way out is through: spend the time on the things you don't know, figure out who will answer your "stupid" questions. Then ask them, even if it's anonymously online
Learn to do your own calculations, even when it's tempting to lean on others
Learn to do your own calculations, even when it's tempting to lean on others
December 4, 2025 at 9:22 PM
I cannot agree with this more, and I'd add that I've never been mad that I spent a lot of time learning something thoroughly — and frequently it does take me a lot of time — whereas I've often been mad (and embarrassed) that I didn't take the time to learn something that I should know.
Reposted by Kyle Cranmer
Since I'm home sick and X-Twitter may soon expect me to pay for access, I am going to recreate this here: 🎢 ⚛️
Pedro Pascal as classic physics textbooks: a thread
Pedro Pascal as classic physics textbooks: a thread
October 31, 2023 at 4:41 PM
Since I'm home sick and X-Twitter may soon expect me to pay for access, I am going to recreate this here: 🎢 ⚛️
Pedro Pascal as classic physics textbooks: a thread
Pedro Pascal as classic physics textbooks: a thread
Reposted by Kyle Cranmer
Tag yourself, I’m the two bathtubs in the laundry room
December 1, 2025 at 9:09 PM
Tag yourself, I’m the two bathtubs in the laundry room
Reposted by Kyle Cranmer
Great read. Too many people assume that the role of theory papers in neuro is to "explain neural data". I'm not even sure we can explain anything yet. Data is more like a muse for theory.
How I contributed to rejecting one of my favorite papers of all times, Yes, I teach it to students daily, and refer to it in lots of papers. Sorry. open.substack.com/pub/kording/...
How I contributed to rejecting one of my favorite papers of all time
I believe we should talk about the mistakes we make.
open.substack.com
December 2, 2025 at 1:52 AM
Great read. Too many people assume that the role of theory papers in neuro is to "explain neural data". I'm not even sure we can explain anything yet. Data is more like a muse for theory.
As we go into the Thanksgiving holiday, I wanted to express my thanks to my collaborators @johannbrehmer.bsky.social @glouppe.bsky.social, Juan Pavez, @smsharma.bsky.social. Recently, I was awarded the Pritzker Prize for AI in Science for work on SBI. That wouldn't have never happened without them.
November 26, 2025 at 6:27 PM
As we go into the Thanksgiving holiday, I wanted to express my thanks to my collaborators @johannbrehmer.bsky.social @glouppe.bsky.social, Juan Pavez, @smsharma.bsky.social. Recently, I was awarded the Pritzker Prize for AI in Science for work on SBI. That wouldn't have never happened without them.
Job🚨: AI & theoretical physics postdoc jointly with @UWMadPhysics & @Perimeter:
* Agentic approaches & test-time scaling methods
* Symbolic verification & tool usage
* LLMs & evolutionary algorithms
* Fine-tuning & RL
* benchmarks tpbench.org
academicjobsonline.org/ajo/jobs/31141
* Agentic approaches & test-time scaling methods
* Symbolic verification & tool usage
* LLMs & evolutionary algorithms
* Fine-tuning & RL
* benchmarks tpbench.org
academicjobsonline.org/ajo/jobs/31141
November 19, 2025 at 1:39 PM
Job🚨: AI & theoretical physics postdoc jointly with @UWMadPhysics & @Perimeter:
* Agentic approaches & test-time scaling methods
* Symbolic verification & tool usage
* LLMs & evolutionary algorithms
* Fine-tuning & RL
* benchmarks tpbench.org
academicjobsonline.org/ajo/jobs/31141
* Agentic approaches & test-time scaling methods
* Symbolic verification & tool usage
* LLMs & evolutionary algorithms
* Fine-tuning & RL
* benchmarks tpbench.org
academicjobsonline.org/ajo/jobs/31141
Reposted by Kyle Cranmer
New location! Zaid Harchaoui's seminar will cover the origins of zero-shot prediction, its latest incarnations in computer vision and language modeling, and current opportunities and challenges. Join us on Nov. 20, 4pm in 2516 Morgridge Hall. dsi.wisc.edu/2025/11/11/d...
November 17, 2025 at 9:46 PM
New location! Zaid Harchaoui's seminar will cover the origins of zero-shot prediction, its latest incarnations in computer vision and language modeling, and current opportunities and challenges. Join us on Nov. 20, 4pm in 2516 Morgridge Hall. dsi.wisc.edu/2025/11/11/d...
On my way to UMichigan for the Annual Data Science & AI Summit at MIDAS
midas.umich.edu/events/midas...
midas.umich.edu/events/midas...
November 17, 2025 at 9:53 PM
On my way to UMichigan for the Annual Data Science & AI Summit at MIDAS
midas.umich.edu/events/midas...
midas.umich.edu/events/midas...
Reposted by Kyle Cranmer
Trump un-endorses Marjorie Taylor Greene
November 15, 2025 at 1:33 AM
Trump un-endorses Marjorie Taylor Greene