Newsletter 21

This is your monthly reminder that at marimo, we’re building the world’s best Python development environment for data — the one that researchers, educators, and engineers deserve. It’s all free and open source: just pip install marimo or uv add marimo to get started!

You’re reading the 21st marimo newsletter.

Some news: we’re hiring! Read on to learn about our open roles. In the rest of the newsletter, we’ll share highlights from 0.19.0 (our biggest release yet), reflect on 2025, and preview what we have in store for 2026.

If you haven’t already, please give us feedback by taking our two-minute survey.

Apply to join marimo team

marimo team is growing! We’re a small team of engineers and researchers who believe the tools we use shape the way we think. If you’re excited about open source, developer tools, or building highly scalable cloud services, please apply!

Open roles. With over 1 million monthly downloads, 18k+ GitHub stars, and adoption at companies, universities, and AI labs worldwide, we’re scaling up to meet the moment. We’re hiring for several roles across our open-source tools and cloud platform:

• Senior Developer Relations Engineer — Growth (SF): In this zero-to-one role, you will harness our users’ love for marimo to build and grow an intentional, structured community of researchers, educators, students, and engineers.
• Staff Engineer, Infrastructure — molab (SF/NY preferred): We are massively scaling up molab, our free hosted notebook service, and we need world-class engineers to help us do it. This role is for a staff engineer, but there is no level cap on this role. We can hire Principal or higher, the more experienced you are the better.
• Senior Engineer, Frontend/Fullstack — Core Open Source (SF/NY preferred): Work across the stack on marimo’s core open-source notebook, library, and ecosystem, doing whatever it takes to make marimo the standard computational notebook across research and industry.

Release 0.19.0

We just shipped marimo 0.19.0, our biggest release yet.

This release simplifies the user interface, paring the sidebar down to the essentials and minimizing visual noise: we’ve lightened shadows, reduced hover styles, and removed layout shifts so you can focus on your data and algorithms without distraction.

For power users, we’ve made the sidebar configurable and have introduced a Developer Panel at the bottom of the editor (toggle with Cmd/Ctrl + J), similar to VS Code’s panel or browser DevTools. This panel houses specialized tools like the terminal, logs, tracing, and scratchpad.

Other highlights:

• pytest fixtures support for easier testing in notebooks, which is especially useful for autograders;
• reactive Plotly bar charts and heatmaps that automatically trigger downstream computation;
• a new dependencies panel that features the minimap for compact notebook structure overview
• parametrizable app composition, making it easier to reuse notebooks as reusable components

Check out our launch blog for more details, and see the changelog for the complete list of 200+ commits.

2025 in review

2025 was an important year for marimo team:

• We grew the team from the two co-founders to a globally distributed team of 8
• We started a YouTube channel and grew it from 0 to nearly 1 million cumulative views by the end of the year
• We launched molab, our free cloud-hosted marimo notebook service
• We made it easy to open notebooks hosted on GitHub, marimo or Jupyter, in molab
• We launched a brand new VS Code/Cursor extension
• We gave talks at PyCon US and SciPy, and met so many amazing people
• We were featured by Nature as a reason scientists should use marimo
• We joined Weights & Biases and CoreWeave to scale molab and build the world’s best developer platform for researchers and engineers
• We shipped, and shipped, and shipped, making more releases than anyone can be expected to keep track of

Where we’re going

We have big plans for 2026. Here are a just a few of the things we’re working on:

• A JupyterHub extension, making it easy to use marimo in existing JupyterHub deployments
• Making it easier to author and use highly scalable, domain-specific anywidgets
• Refreshed slides and grid layouts
• Fully featured IPython notebook exports
• PDF export
• Revamped Quarto extension, letting you use marimo more easily in Posit products
• Remote storage inspector, making it easy to work with object storage, Google Drive, and more
• Debugging in VS Code
• molab secrets
• molab instant previews, letting you share snapshots of your work
• a Kubernetes operator

Want us to prioritize something not on this list? Let us know on Discord or GitHub.

🍃 Community

We’re at 180k+ weekly downloads, 18k+ stars on GitHub, 200+ contributors pushing code to marimo, 940k+ YouTube views, and 3k+ marimonauts hanging out on Discord — join the conversation!

2025 Roundup. 2025 was the year that our community really took off. Here are some uses and mentions of marimo in the wild that we found particularly inspiring.

• marimo in security engineering. Matt Franz explains why threat hunters should stop using Jupyter and switch to marimo
• Interactive learning. Learn how to implement GRPO, an RL algorithm used by DeepSeek R1, in the Hugging Face NLP course that features inline marimo notebooks
• A Real Python course. Real Python, the world’s best site for Python education, launched a comprehensive marimo course.
• Nature features marimo. Nature, the world’s most renowned science journal, highlights marimo and its emphasis on reproducibility and reusability as a reason that computational scientists should use Python.
• Cloudflare bets on marimo. Cloudflare announced they’re betting on marimo as the interface to Cloudflare data, launching notebooks.cloudflare.com.
• Production-ready data science. Khuyen Tran’s new book Production-ready data science features marimo as the solution for reproducible notebooks
• Agentic visual reporting with marimo notebooks as artifacts. Péter Gyarmati won 1st place at IEEE VIS conference’s VISxGenAI workshop with Agentic Visual Reporting, featuring marimo notebooks as artifacts
• A Kaggle Grandmaster makes the switch to marimo from Jupyter. Kaggle Grandmaster Parul Pandey explains why she made the switch to marimo notebooks

Don’t forget to submit your projects to our awesome-marimo repo!

Sincerely,

marimo team 🍃