Akshay Agrawal
@akshaykagrawal
Building @marimo_io. PyMDE, CVXPY, & ex TensorFlow developer. Minimum-Distortion Embedding book author. Stanford BS/MS/PhD.
My PhD thesis was a 100+ page monograph on vector embeddings. When I worked on it, I really wanted embeddings to be tangible: to select points in a 2D embedding, and get the original data back in Python. I couldn't do that in Jupyter ... but you can do it today in marimo.
marimo brings your data to life! Select points in a plot and get them back in Python, instantly, with reactive charts. In this data app, a reactive chart lets the user interactively explore the original images backing a 2D embedding of MNIST.
OMG! New low for NeurIPS reviews. What a travesty journal and conference peer review has become for different reasons. The whole system is broken & needs a complete revamp.
Anyone knows adam?
Building molab was an all-hands-on-deck sprint. Really proud of this launch and how quickly it came together! I'm also really proud that we did all this while also shipping some huge OSS releases at the same time.
Announcing molab: a cloud-hosted marimo notebook workspace with link-based sharing. Experiment on AI, ML and data using the world’s best Python (and SQL!) notebook. Launching with examples from @huggingface, @weights_biases, and using @PyTorch marimo.io/blog/announcin…
In the spirit of building in public, here's our first round of evals for our "Tab Tab" model we are training for @marimo_io notebooks. The task is fill in the middle, cursor-like code completion and cursor prediction. Looks like for this task Qwen > Llama, and we're already…
Come hang with the @marimo_io community on August 16! 🍃
We're having our community call on August 16, 2025 from 10:30-11:30 AM PST. Join us as we discuss the latest updates, highlight community projects, and more! RSVP: lu.ma/ifs1nwzk
Brought to you by the one and only @trevmanz
Reactive execution sets marimo apart from traditional notebooks. It keeps code and outputs in sync and prevents bugs before they happen. Our latest release includes an interactive minimap that helps you understand how data flows through your notebook.
observable-like minimap coming soon to a @marimo_io notebook near you 🫡 navigate and visualize the relationships between cells:
The Modal Python SDK now has first-class support for uv, including both `uv_sync` and `uv_pip_install` 🤝
👇A @marimo_io notebook in an official @PyTorch repo 🍃
To show how to use the new debug flag I whipped a lil marimo notebook, mostly because @marimo_io has amazing youtube videos that made me want to explore! It can be found here: github.com/pytorch-labs/a…
As seen on the front page of HN ... ! 👀
Announcing molab: a cloud-hosted marimo notebook workspace with link-based sharing. Experiment on AI, ML and data using the world’s best Python (and SQL!) notebook. Launching with examples from @huggingface, @weights_biases, and using @PyTorch marimo.io/blog/announcin…
apropos of nothing: this is an appreciation post for @simonw, who tries out literally all the tools, and joyfully shares (for free!) what he learns in the hope that it will be useful to other humans grateful to have someone relentlessly building with the garage door open 💗🙌
That week in June was pretty epic: from the SF Systems reading group on Firecracker VMs, to jamming with @ekzhang1 on @marimo_io on @modal_labs, at the @aixventureshq, to molab today!
the story was, in June, I was so excited about a "Marimo cloud" that I went to their office & we hacked together the first prototype of a Modal Sandbox backend for Marimo in a couple hours. just a month later, guess they have a whole new cloud-based product, still on Modal :D
marimo offers a reactive environment for seamless reproduction and sharing! Enjoy auto cell updates and correct execution. Use the link below 👇 to discover how marimo’s features make it an ideal tool for documenting research and learning activities.
Proud of this milestone: @CloudflareDev bets on @marimo_io as _the_ interactive programming environment to explore @Cloudflare data with Python and SQL
It turns out that one of the most popular feature requests in Google Colab’s issue tracker is: support marimo. Rather than wait for wait for Google to act ... we decided to just build molab instead.
I haven't been using @marimo_io for long but I can say I'm already totally hooked. Auto-updating cells is amazing for discovery.
Proud of this milestone: @CloudflareDev bets on @marimo_io as _the_ interactive programming environment to explore @Cloudflare data with Python and SQL
This was an exciting collaboration (led from our side by @themylesfiles!). We're also launching with many examples for working with data on R2, workers AI, and more.
We’ve partnered with marimo to bring their best-in-class Python notebook experience to your Cloudflare data: cfl.re/4eRBmMU
. @RealPython has very high quality tutorials (such as their one on asyncio). Excited to see they have a @marimo_io video course!
📺🐍 Getting Started With marimo Notebooks realpython.com/courses/gettin…
We heard that some of you like to claude code ... ... so we think you'll like this feature, --watch this! youtu.be/tOyZM7HswSk
We're doing a livestream later today! This time we're explore @pydantic's AI library with @marcelotryle. You can join live below, or rewatch later on the same link: youtube.com/watch?v=ujQjqq…