iiinova — where your digital self keeps evolving

Reposted by nathan

Aishik Ghosh @aishikghosh.bsky.social · 10/11/2025

📢𝐉𝐨𝐛 𝐚𝐥𝐞𝐫𝐭: Postdoc position(s) in 𝐀𝐈 𝐟𝐨𝐫 𝐟𝐮𝐧𝐝𝐚𝐦𝐞𝐧𝐭𝐚𝐥 𝐩𝐡𝐲𝐬𝐢𝐜𝐬 focusing on either 𝐬𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐚𝐥 𝐢𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞 or 𝐀𝐈-𝐚𝐬𝐬𝐢𝐬𝐭𝐞𝐝 𝐭𝐡𝐞𝐨𝐫𝐲. 🏙️Join my new group at 𝐆𝐞𝐨𝐫𝐠𝐢𝐚 𝐓𝐞𝐜𝐡, based in 𝐰𝐨𝐧𝐝𝐞𝐫𝐟𝐮𝐥 𝐌𝐢𝐝𝐭𝐨𝐰𝐧 𝐀𝐭𝐥𝐚𝐧𝐭𝐚! 🗓️ Reviews begin sooner, deadline: 𝟓𝐭𝐡 𝐃𝐞𝐜𝐞𝐦𝐛𝐞𝐫 💬 Questions: Reach out! 🔗 inspirehep.net/jobs/2962185

inspirehep.net

INSPIRE

042

nathan @nds.bsky.social · 05/11/2025

i deeply need the video essay on the typographical success of the Zohran campaign

010

Reposted by nathan

Lukas Heinrich @lukasheinrich.com · 31/10/2025

🚨 Job Alert We're hiring our first Postdoc to work with us on our @erc.europa.eu Project on Generative AI for Particle Physics. Check out the details below and don't hesitate to get in touch or spread the word! Deadline: Nov 30th inspirehep.net/jobs/3075448

lnkd.in

This link will take you to a page that’s not on LinkedIn

0119

Reposted by nathan

Aki Vehtari @avehtari.bsky.social · 03/11/2025

Now I'm also looking for a research software engineer to implement a pile of research results to R packages loo, posterior, bayesplot, projpred, priorsense, brms or/and Python packages ArviZ, Bambi and Kulprit. Apply by email with no specific deadline (see contact info at users.aalto.fi/~ave/)

25551

Reposted by nathan

Trans+ Solidarity Alliance @transsolidarity.bsky.social · 15/10/2025

URGENT ACTION 🚨 If you do one thing for trans rights today – contact your MP about the proposed EHRC Code of Practice, which would devastate trans people's lives. Even if you've sent a letter before, make sure you use our new tool to send another. Now is the moment to keep the pressure up 👇

actionnetwork.org

Scrap the Bathroom Ban

🚨This is urgent🚨 Bridget Phillipson, the Minister for Women and Equalities, has a decision to make. The EHRC, a regulator mired in controversy over anti-trans views, has sent a new draft Code of Prac...

15539

Reposted by nathan

Carl Zimmer @carlzimmer.com · 09/10/2025

Today my @nytimes.com colleagues and I are launching a new series called Lost Science. We interview US scientists who can no longer discover something new about our world, thanks to this year‘s cuts. Here is my first interview with a scientist who studied bees and fires. Gift link: nyti.ms/3IWXbiE

nyti.ms

14147041812

nathan @nds.bsky.social · 15/09/2025

just wanted to share that after nearly 8 years of programming in Python and dabbling with other languages as-needed, I've finally written my first bash script! having random blind spots as professional does not make you bad at your job, and i just wanted to normalize that :)

110

Reposted by nathan

Rajni Shah (they, them) 🍉 @rajnishah.bsky.social · 26/07/2025

"You might feel powerless. That it is too late. It is way too late for way too many. But way too late is going to have to be soon enough." 'The Sound of Solidarity' from @saranahmed.bsky.social feministkilljoys.substack.com/p/the-sound-...

feministkilljoys.substack.com

The Sound of Solidarity

“From Birzeit University, in the heart of occupied Palestine, we are screaming with Gaza.

1159

nathan @nds.bsky.social · 22/07/2025

please use this fantastic email tool to make some noise about the EHRC Code of Practice!! some awful stuff might be made law without a parliamentary vote. different links for both trans people and allies :) www.scottishtrans.org/emailfortransinclusion

scottishtrans.org

Email your MP and Minister for Equalities about trans people's need for safe inclusive services - Scottish Trans

000

nathan @nds.bsky.social · 03/06/2025

with how good AI-assisted coding is now, are programmers going to become more of a luxury trade? "We hand-craft artisanal Python code to solve your use-case. Experts weave every letter of the variables into your codebase. All commits are done manually -- personalize your messages for a small fee."

000

nathan @nds.bsky.social · 26/05/2025

obsessed with this vinted account that models all their items with a djungelskog

010

Reposted by nathan

Simon Willison @simonwillison.net · 22/05/2025

If your library doesn't have any documentation, it can't have any bugs simonwillison.net/2025/May/22/...

If your library doesn't have any documentation, it can't have any bugs.

Documentation specifies what your code is supposed to do. Your tests specify what it actually does.

Bugs exist when your test-enforced implementation fails to match the behavior described in your documentation. Without documentation a bug is just undefined behavior.

If you aim to follow semantic versioning you bump your major version when you release a backwards incompatible change. Such changes cannot exist if your code is not comprehensively documented!

Inspired by a half-remembered conversation I had with Tom Insam many years ago.

910618

nathan @nds.bsky.social · 18/05/2025

being 194cm, this has always been the most frustrating thing about living & travelling in the uk compared to the rest of europe -- i have to go out of my way to get stupidly wide beds just to increase the length & not have my feet dangle off the end 😭

Chart showing bed sizes in the UK -- the euro single notably being larger than UK single.

110

Reposted by nathan

Charlie Marsh @crmarsh.com · 13/05/2025

Today, we’re announcing the preview release of ty, an extremely fast type checker and language server for Python, written in Rust. In early testing, it's 10x, 50x, even 100x faster than existing type checkers. (We've seen >600x speed-ups over Mypy in some real-world projects.)

1433284

Reposted by nathan

Good Law Project @goodlawproject.org · 03/05/2025

We believe the Supreme Court has placed the UK in breach of its obligations under the Human Rights Act. Over 8,000 people have supported our case - will you join us? goodlawproject.org/crowdfunder/...

goodlawproject.org

Help us challenge the Supreme Court’s judgment on trans rights | Good Law Project

We use cookies where necessary to allow us to understand how people interact with our website and content, so that we can continue to improve our service.

123390

nathan @nds.bsky.social · 05/05/2025

well, this is a new one for me 😄 turns out that python's list creation shorthand for duplicating elements (e.g. x = [obj] * n) does not make the instances of obj distinct in memory! TIL.

010

Reposted by nathan

Gilles Louppe @glouppe.bsky.social · 29/04/2025

<proud advisor> Hot off the arXiv! 🦬 "Appa: Bending Weather Dynamics with Latent Diffusion Models for Global Data Assimilation" 🌍 Appa is our novel 1.5B-parameter probabilistic weather model that unifies reanalysis, filtering, and forecasting in a single framework. A thread 🧵

25215

nathan @nds.bsky.social · 22/04/2025

looking into distrax a bit to implement some normalising flow stuff for a talk — found out the CI is failing because pylint flags the *whole library* for bad indentation (google uses 2-space tabs). this is what happens when you make double standards 😌 github.com/google-deepm...

github.com

tests · google-deepmind/distrax@efeeee8

Contribute to google-deepmind/distrax development by creating an account on GitHub.

000

Reposted by nathan

Elise (they/them)🇦🇺 🏳️‍⚧️ @eliseswords.eurosky.social · 16/04/2025

So sorry trans cousins in the UK. I'll never forgive that author, the author of the notorious report or associated 'adult human female' shitheads for the hatred they've spread against you. This "lets hate trans people instead of doing anything to fix anything" bullshit needs to STOP.

1203

Reposted by nathan

Peyman Milanfar @docmilanfar.bsky.social · 11/04/2025

bias-variance tradeoff

0545

nathan @nds.bsky.social · 03/04/2025

idk if this is a neurodivergent thing, but having a low-volume audiobed with some kind of background music can really help me stay in long meetings when i'm remote. i'd recommend giving it a try if your concentration can waver easily — i find it surprisingly grounding!

000

nathan @nds.bsky.social · 13/03/2025

yay im a real ml engineer now 📉

000

Reposted by nathan

Adriana Guatame-García @adriguatame.bsky.social · 06/03/2025

A friend included this anti-acknowledgement section on her PhD thesis. She also added the proposition: “Systematic bullying and undermining of girls and women in STEM starts early on and is the reason why they do not stay in science and related fields.” Absurd we still need to go through this 🧪👩‍🔬

There have been people less than helpful in my journey here. I wanted to acknowledge those too, bc I know I am not unique in this experience.
No thank you to the physics study assoc that made me sing songs about how women couldn't study physics without sleeping with the professor, the day I stepped into university life. No thank you to the 5th year physics student that decided to assign me a stripper name within the first minute of meeting me in the physics coffee corner in my first year. No thank you to the technician that was responsible for onboarding me on the use of the cluster in my third year who raised his eyebrows and asked me if that meant I was some sort of computer girl. No thank you to the senior researcher that sent me utterly inappropriate texts after a conference, then proceeded to apologise months later by telling me they had not been meant for me anyway so no hard feelings remain hopefully And no thank you to him for attending every conference I've been to since. No thank you to the people who told me that it was surprising that I was doing a PhD since I was a girl. No thank you to the man who mistook me for a coffee lady at a conference, and after having to correct him two times that I did not work there, responded with you should consider it. No thank you to the researcher that asked me what I was wearing underneath my outfit during a conference. No thank you to the physicist who declared to a room full of other physicists that biologists don't know how to design an experiment. No thank you to the people who have called me scary instead of strong and intimidating instead of intelligent.
And finally, no thank you to the exec board of the TU Delft, whose knee-jerk reaction to being held up a mirror about the social safety at the university, was to sue the party holding up the mirror instead of looking at the problems they highlighted.
... You have made me feel like I do not belong in science & I cannot forgive you for that.

12365132320

Reposted by nathan

Randy Chase @dopplerchase.bsky.social · 25/02/2025

Mark today in the history books. The first AI weather prediction model labeled “operational” by a major world forecast center. Congrats ECMWF 🎉

0277

nathan @nds.bsky.social · 25/02/2025

after enjoying some testing of @simonwillison.net's LLM CLI tool, going to try some more regular use of models for my more organisation-related tasks. what model would people suggest that meets these reqs: - long context (e.g. many long text files) - economic - if local, fits in 32GB mem?

100

nathan @nds.bsky.social · 09/02/2025

the em-dash is a primary love language of mine — just look at all the mental space that made! and our context is perfectly preserved 😌

010

Reposted by nathan

Gilles Louppe @glouppe.bsky.social · 07/02/2025

My take: AI will make us artificially intelligent.

391

nathan @nds.bsky.social · 05/02/2025

this is heartbreaking. people’s stories are being censored — literally erasing records of their history and impact live in front of our eyes. now is the time to *scream* about your colleagues, mentors, friends. uplift the voices of those that this system is targeting. make them undeniable.

021

Reposted by nathan

Simon Willison @simonwillison.net · 04/02/2025

Tiny TIL: I just figured out how to run pytest with a different Python version against my pyproject.toml/setup.py projects using uv run uv run --python 3.12 --with '.[test]' pytest til.simonwillison.net/pytest/pytes...

til.simonwillison.net

Running pytest against a specific Python version with uv run

While working on this issue I figured out a neat pattern for running the tests for my project locally against a specific Python version using uv run :

710414

Reposted by nathan

Floriane Magera @flomgr.bsky.social · 16/01/2025

I wish I stumbled on this pretty and clear intro to CUDA and GPUs years ago : modal.com/gpu-glossary

1336

Reposted by nathan

Stephan Hoyer @handle.invalid · 09/01/2025

This is a huge milestone for cloud-native big scientific data!

0234

Reposted by nathan

Matthew Feickert @matthewfeickert.com · 07/01/2025

The #SciPy2025 Call for Proposals is live (as of mid December 2024, sorry forgot to share) over on scipy2025.scipy.org! Deadline for submissions is 2025-02-26, so get submitting!

12116

Reposted by nathan

Miles Cranmer @milescranmer.bsky.social · 02/01/2025

🚨 FINAL REMINDER 🚨: Multiple Postdoc and PhD positions in our AI + Physics cluster in DAMTP, Cambridge! Deadlines: - Postdoc: Jan 5th (Sunday) - PhD: Jan 7th (Tuesday) More info below – Please share with researchers and students who might be interested in joining us!

0165

nathan @nds.bsky.social · 30/12/2024

i'm only drip-feeding myself ai-assisted coding through talking to claude sonnet now and again (as opposed to integrated solutions on the scale of cursor to copilot). main reason is the fear of losing the joy of coding! i know they're useful; do people also find these kinds of tools joyful?

000

Reposted by nathan

Gilles Louppe @glouppe.bsky.social · 30/12/2024

📣 Hiring! I am looking for PhD/postdoc candidates to work on foundation models for science at @ULiege, with a special focus on weather and climate systems. 🌏 Three positions are open around deep learning, physics-informed FMs and inverse problems with FMs.

47934

Reposted by nathan

Stephan Hoyer @handle.invalid · 19/12/2024

Can incorporating AI improve precipitation in global weather and climate models? Yes! In the latest NeuralGCM paper, we show that training on satellite-based precipitation results in significant improvements over traditional atmospheric models: arxiv.org/abs/2412.11973

arxiv.org

Neural general circulation models optimized to predict satellite-based precipitation observations

Climate models struggle to accurately simulate precipitation, particularly extremes and the diurnal cycle. Here, we present a hybrid model that is trained directly on satellite-based precipitation obs...

1345

Reposted by nathan

Jeremy Howard @howard.fm · 19/12/2024

I'll get straight to the point. We trained 2 new models. Like BERT, but modern. ModernBERT. Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff. It's much faster, more accurate, longer context, and more useful. 🧵

19620147

Reposted by nathan

Margaret Mitchell @mmitchell.bsky.social · 11/12/2024

Inspired by @wimlds.bsky.social , I looked for a "Women in Machine Learning" starter pack and couldn't find one. So I created one! May have some mistakes. I'll try to do an AI ethics one next. 🤗 go.bsky.app/LT6CwNN

53406136

nathan @nds.bsky.social · 05/12/2024

reading detailed issues is one of my favourite ways to learn about software. it can have the side-effect of making me trust the corresponding package more due to the kind & helpful intent shown by the devs/contributors. i think this is also why i like trying to be thoughtful with issues myself :)

010

nathan @nds.bsky.social · 03/12/2024

claude sonnet 3.5 holding no punches

000

Reposted by nathan

nate @zzstoatzz.io · 21/11/2024

uv makes it so easy to go zero to MRE say i find bug in foo==x.y.z, i can repro with: uv run \ --with foo==x.y.z \ repro.py RuntimeError: oh no! test the branch with a fix: uv run \ --with foo@git+https://{giturl}.git@hotfix-foo \ repro.py ✅ just used this today! github.com/pydantic/pyd...

github.com

new requirement in 2.10 to pass `validated_data` when `call_default_factory=True` · Issue #10912 · pydantic/pydantic

Initial Checks I confirm that I'm using Pydantic V2 this issue opened per #10905 (comment) Description appears to be a new requirement that validated_data is passed to Field.get_default if call_def...

38015

nathan @nds.bsky.social · 21/11/2024

using my niche subculture skills, i attempted a ml for hep starter pack. please say if you notice glaring omissions, it’s nowhere near exhaustive — plus, not everyone is over here yet! go.bsky.app/2E66DfJ

010