Marimo Emerges From Stealth To Debut Its Python Notebook
Marimo has emerged from stealth with $5 million in seed funding. The startup has launched an open-source Python notebook designed to provide developers with a tool that is reproducible and Git-friendly. It can also be deployed for production as a web app and executed as a script.
The seed funding round was led by Anthony Goldbloom (ex-Kaggle, Sumble) and Shyam Mani of AIX Ventures, with participation from Jeff Dean (Google), Clement Delangue (HuggingFace), and other notable investors.
Python has been the dominant language in AI. In October 2024, it overtook JavaScript as the most used language in GitHub. As the lingua franca of AI, machine learning (ML), and STEM, Python’s adoption has continued to expand as these technologies mature and transition into production environments.
To support this transition, several venture-backed startups are modernizing key aspects of the Python stack. MotherDuck and Polars are building fast OLAP engines to replace Pandas. Astral and Prefix are modernizing package management by replacing legacy installers like pip with UV and Pixi. Modal and Coiled are also working to enable serverless functions for use at any scale.
While there has been significant modernization of Python tools to support AI, ML, and data science, the Python notebook, which serves as the primary environment for much of this work, has received little attention.
Marimo plans to fix this with its Python notebook addressing the key shortcomings of traditional notebooks like Jupyter. Marimo claims that, unlike Jupyter, the marimo notebook ensures true reproducibility by keeping track of dependencies and execution flow automatically. This can help eliminate common issues of Python notebooks such as broken code from missing dependencies, difficulties with version control, and the lack of reproducibility.
According to Marimo, their notebook not only addresses the key limitations of traditional tools like Jupyter but also outperforms platforms such as Streamlit and Gradio. The marimo notebook comes with an AI assistant aware of dataframe schemas and integration with commercial coding copilots like GitHub Copilot and Codeium.
Additionally, marimo aims to provide a more user-friendly and expressive environment throughout the lifecycle, from deployment to production. It also comes with a large library of UI widgets like data frame transformers and cheat elements. This approach addresses the limitations of static and error-prone notebooks by enhancing interactivity.
“When I started working on marimo, I had just finished a PhD at Stanford where I helped grow an open-source library for optimization to a million monthly downloads, and before that worked on TensorFlow at Google Brain,” shared Akshay Agrawal, Founder and CEO, Marimo Inc., via a blog post on the seed announcement. “Jupyter notebooks were essential to my work, because they let me see my data while I worked on it — but they were also extremely frustrating.”
“Hidden state led to hidden bugs (over a third of the 10 million Jupyter notebooks on GitHub fail to reproduce), the JSON file format made them hard to use in Python codebases, and the final documents lacked interactivity. While Jupyter notebooks are widely used for AI/ML development, STEM, and data engineering, now more than ever, there’s growing consensus that this kind of work shouldn’t be done in error-prone scratchpads.”
Akshay, along with co-founder Myles Scolnick, wanted to create a programming environment that “blended the best parts of interactive computing with the rigor of traditional software development.”
Within a year of its launch, Marimo has been downloaded over 1 million times. It is being used at several organizations and universities around the globe, including BlackRock, Mozilla AI, MLB, and Stanford. The online community is also growing strong, with over 7.8k stars on Github, 60+ contributors, and 1k members on Discord.
Marimo plans to use the funds to continue developing the Marimo notebook and explore creating other data tools. As the company focuses on refining the notebook, they do have plans to launch a paid product in the future. However, the company remains committed to open-source development, ensuring its tools are free and accessible.
Related Items
Anaconda’s New Tool Lets Users Run Python Code Inside Excel
Python Now a First-Class Language on Spark, Databricks Says
KX Aims to Unlock Python’s Potential with PyKX 3.0