Sean Arms

@lesserwhirls.bsky.social

☦️ Indiana born, PhD in Meteorology from OU, software engineer at NSF Unidata (UCAR/UCP), priv/acc

What questions do you have about AI in education? 2026 NSF Unidata Summer Intern Tara Staresinic has published the first post in her "AI in Education" series, sharing a student's perspective on AI, sustainability, ethics, and practical ideas for the classroom. buff.ly/BhPXGMT

Bild

🎉Python Package Milestone🎉 Proud to announce that xmacis2py reached 10,000 total downloads on conda-forge as of today. xmacis2py is a Python interface with the ACIS2 API for retrieving climate data and has lots of tools to analyze and visualize the data. xmacis2py = xmACIS2 in Python. #climate

BildBildBildBild

If I were the King of Universities (which, heaven forbid), I would place this finding at the exact centre of every single policy decision from teaching to research to estates to capital investment. At every level in every process, the question would be: are we as humane as education demands we be?

Screenshot from a Higher Education Policy Institute report. The key sentence is: "students are used to and want human teaching, human feedback, and human connection."

🎉xmACIS2Py 2.5.1 Released🎉 Users can now calculate daily departures from normal based on a custom climatology in xmACIS2Py 2.5.1. xmACIS2Py 2.5.1 brings two new Jupyter Notebook Tutorials where we make ENSO Temperature & Precipitation Analog Composites. Notebook Links👇 #python #climate #weather

BildBildBildBild

MetPy Mondays is back! John Leeman from NSF Unidata introduces what’s new in MetPy and previews a new 15-part MetPy Mondays series focused on practical meteorological data analysis, remote data access, and case study plotting. buff.ly/BQitB7I

MetPy Mondays #329 - What's New in MetPy? | NSF Unidata

This MetPy Monday episode introduces a new series focused on the latest features in MetPy and practical workflows for meteorological data analysis in Python. John highlights improvements to core…

unidata.ucar.edu

"...If you hand that process to a machine you haven't accelerated science. You've removed the only part of it that anyone actually needed." I hear all the time about the importance of efficiency in so many aspects of our lives. But HUMANS ARE NOT MACHINES. Learning & growth is slow & a GOOD thing!

I'm personally banking (as an old fuck) that being able to write actual code that actually works will be as valuable in a few years as old fucks that still remembered how to write COBOL around Y2K

watching GitHub slip to one-nine of uptime over the last month or so has been an interesting reminder that LLM generated code is like taking out a high interest loan against your future self; there are ways of leveraging that debt into something valuable but most of the time, you default

I’ve said it before, but this time more visibly (on National AI Literacy Day nonetheless). For your consideration: Changes that encourage students, or scientists, to move further away from the data should be violently challenged. 🔥 (╯°□°)╯︵ 🔥┻━┻🔥 #AILiteracy

UCS filed Freedom of Information Act requests to the NSF to publicly release all of its communications about its plans to restructure the National Center for Atmospheric Research (NCAR). “NCAR represents one of the most successful models of federally supported scientific infrastructure in the US.”

UCS Files Requests to Make Proposals Public on How to Restructure National Center for Atmospheric Research

The Union of Concerned Scientists today filed Freedom of Information Act requests to the U.S. National Science Foundation to publicly release all of its communications about its plans to restructure t...

ucs.org