DEVOPSdigest

Bauplan, a Python-first serverless data platform that transforms complex infrastructure processes into a few lines of code over data lakes, announced its launch with $7.5 million in seed funding.

The round was led by Innovation Endeavors with participation from experienced operators Wes McKinney, Aditya Agarwal, Chris Re. Ihab Ilyas joins as an advisor.

Bauplan offers a serverless runtime that processes large datasets directly on object storage in pure Python. Developers can build powerful applications using simple serverless Python functions and familiar git-like concepts like branch, commit, and merge. This programmable, code-first platform is built for automation to empower the next generation of data developers and eliminates the need for Kubernetes, Spark, and special infrastructure management.

“Data today looks a lot like DevOps a decade ago,” said Ciro Greco, CEO and co-founder of Bauplan. “Back then, infrastructure-as-code allowed all kinds of developers to automate a lot of stuff. Data is going through the same process today. We had a revelatory moment at the beginning of this year when a large infrastructure team put the system in production and we went from zero to 40,000 jobs per week.”

Bauplan was created by Greco, Jacopo Tagliabue, and Mattia Pavoni, who took their previous company, Tooso, from inception to IPO through acquisition by Coveo. The team has worked on data and machine learning for a long time, published 60+ research papers, earned thousands of citations, and created popular open source contributions with over 50M downloads and 10k GitHub stars.

Bauplan is ideal for infrastructure and data science teams in medium sized and large enterprises with data intensive use cases involving machine learning and AI applications, especially in B2B software, media, financial services and healthcare tech. It’s already used by enterprise design partners, such as MFE-MediaForEurope, a leading European broadcaster.

The new capital will be used for product development and initial market validation with early customers.

The platform provides serverless Python functions vertically integrated with object storage, supporting natively Iceberg tables and git-like operations over data lakes, like zero-copy branches and automatic data versioning. Developers can build pipelines and applications over data branches without having to manage Kubernetes, Spark, or any other infrastructure and manage the entire data lifecycle as a CI/CD workflow using a simple Python SDK.