A cool idea from a colleague: a claude/cursor/etc plugin that searches for SKILL.md files in the installed packages of your Python environment

Except maybe using conda's env (de-)activation mechanism to run a small script that places skills in the right place

This looks nice! Especially the built in versioning.

A downside is that it is "yet another thing to install", it isn't enough to just install the original package. But I can't think of a great way of achieving the "one package" dream (without extra commands to run)

How do you distribute AI Agent skills to users of your python packages? As in, how do people who `pip install` your work get a skill that is relevant to your project?

So war der Winter 2025/26: Ungünstiger Schneedeckenaufbau, sehr grosse Lawinengefahr im Februar, viele Lawinenunfälle
Alle Infos gibt's im Winterrückblick: www.slf.ch/de/lawinenbu...

📷 M. Senn
1/3

nteract 2.0 A ground-up rebuild of the notebook app. New runtime, new architecture, new everything.

Announcing nteract 2.0! www.nteract.io/blog/nteract...

Native notebook app with managed python environments, realtime collaboration with agents, and some @deno.land to boot.

Silly agent. Asked for a happy little plot and it delivered.

How do you ship AI Agent Skills with your Python library? That is skills for the users of your library, not developers

The next scikit-learn release will allow inspecting the type and values of attributes of fitted estimators in Jupyter notebooks & example code rendered as HTML in sphinx-gallery powered project websites.

scikit-learn.org/dev/auto_exa...

Why do some people love AI for building software and others hate it? My guess is that people who happen to use code to solve a problem love AI, those who like the process of crafting code hate it.

WDYT?

Infrastructure Engineer - Python Software Foundation - Career Page Apply to Infrastructure Engineer at Python Software Foundation in Remote.

come work with meeee 🥹

The Python Software Foundation is hiring an infrastructure engineer.

Apply: pythonsoftwarefoundation.applytojob.com/apply/DNzZlB...

yeah, but I worked so hard on that new turn signal. I need you to know about it!!

A snowy mountain ridge that has a few tracks from skiers going along it on it. Bright blue sky.

View across a snowed in plateau with alps in the background

View of mountains with ski touring tracks in the snow. Big, low clouds in the background. Blue sky above

Weekend views

So how does it work? Meet Miniature (mini + IA, get it?), a complete AI agent in ~225 lines of Python. No frameworks, no abstractions. Code: github.com/glouppe/miniature

For the last two day Strava hasn't manage to sync activities from Garmin. It feels like the early days of Strava again :D

How far away are we from the point where humans only use issues and PR comments, bots act on the issues and make PRs that then get human review and more input

Instead of using an editor I might as well use the GitHub UI directly no?

Thoughts on using AI in public: betatim.github.io/posts/using-...

How do you use AI tools to work on a project as a team?

The day my coffee and milk unmix is the day a backlog will get longer. That "Publish paper on how entropy is bunk" project won't get done by itself ;)

The age old cycling advice for fixing problems: ride harder!

Cold fingers? Ride harder!
Not fit enough? Ride harder!
Miserable rainy ride? Ride harder!
AI not letting your focus on riding? Ride harder

Have fun out there on the bike. Claude will be waiting when you get back

I just made a sidecar viewer of output you can run in jupyter console or even a regular ipython terminal to get rich interactives without a notebook.

pip install runtimed

import runtimed
runtimed.sidecar()

Seriously, shall we still teach students how to even program?

A bit like the skill that a scientist can tell if a scientist knows what they are talking about or not. Even if it isn't their field of research. You can't suggest the right solution, but you can tell that it is not quite right. This is what we need for the future

A little bit. In the (near) future it'll be more important to be able to judge the work of a programmer (e.g. AI tool) than to write code yourself. Learning a bit of programming probably helps with that.

The Synthetic Senior | FOSDEM 2026 Thank you to the organizers: Laura & Shirley The Synthetic Senior FOSDEM 2026 Community Dev Room Abigail Cabunoc Mayes GitHub, SustainOSS, OpenJS @abbycabs Rethinking Free Software Mentorship in the A...

"The Synthetic Senior" — my #FOSDEM talk on AI + open source mentorship

The 3 C's for finding contributors worth investing in:
💡 Comprehension
⚙️ Context
🔄 Continuity

Slides: docs.google.com/presentation...

You might check out some of the resources from Scientific Python if you haven't already:
* learn.scientific-python.org/development/...
* lectures.scientific-python.org/advanced/int...

Question: I want to practice writing functions in C, then bring those functions into a Python package via ctypes. What is the best way to build the project since I am on a Windows OS and thus my compiler is different from Mac and Linux? #python #c #programming

LLMs are always confident about being able to help you, complete a task or answer a question.

They are excellent at hiding their imposter syndrome 😂

In scikit-learn we have been experimenting github.com/scikit-learn... (asks for disclosure in a simple way) as well as github.com/scikit-learn... to try and improve the output when someone uses AI. github.com/scikit-learn... tries to improve the AGENTS.md wrt how to choose issues to work on

Looking for examples: do you know any OSS projects that have thoughtfully updated their contribution process for AI-assisted PRs?

Interested in:
* PR templates asking for intent/reasoning
* Comprehension checks
* Other friction-adding methods

Would love to learn from communities experimenting here