Notebooks are for Production ML Now
There's a great pattern for marimo/Pydantic/PyTorch that more people should know about. You can be creative, and still have your production train job without making a sacrifice to your workflow. And you even get solid weights and biases support for free! 00:00 New pattern 00:59 Features 03:00 Python File 04:52 One big recommendation 06:30 Thing on top 08:08 Skills Relevant links: https://github.com/marimo-team/skills https://github.com/marimo-team/skills... Links: Website: https://marimo.io Discord: https://marimo.io/discord Reddit: / marimo_notebook Twitter: https://x.com/@marimo_io Tiktok: / marimo.io Instagram: / marimo_io Bluesky: https://bsky.app/profile/marimo.io Newsletter: https://marimo.io/newsletter

▶︎
Why Prefect and marimo are perfect together

▶︎
How Notion Cut Millions from Their Vector DB Bill Without Sacrificing Search Quality

▶︎
Android 17 sucks. So I put Linux on a phone.

▶︎
VSCode notebooks just got better

▶︎
AI buys robot and car, does exactly what experts warned.

▶︎
All your notebooks can be reproducible now thanks to UV

▶︎
Why Pydantic Logfire is Great for Batch Jobs & Python Notebooks Too!

▶︎
Vibe coding an ML pipeline: it works now!

▶︎
Introducing: the marimo-quarto plugin!

▶︎
Learn Machine Learning Like a GENIUS and Not Waste Time

▶︎
I Think They Are Lying To You

▶︎
Why Everyone's Using Marimo Operators in Airflow Now

▶︎
How marimo notebooks can be used in four different ways

▶︎
Vibe Coding With Kimi K2.6

▶︎
Every Way To Run Open Source AI Models

▶︎
What are RLVR environments for LLMs? | Policy - Rollouts - Rubrics

▶︎
I tested 3 local AI models. The smallest one won.

▶︎
17 marimo features to boost Python productivity

▶︎
LLM Settings That Every AI Developer Must Know

▶︎
