DVC

DVC (Data Version Control) is an open-source tool, now maintained under the treeverse GitHub organization and branded "DVC by lakeFS", that applies a Git-like model to data, AI/ML and data science work. The repository describes it as "Data Versioning and ML Experiments" and lists version 3.67.1 as the latest release, published 2026-03-31. It is released under the Apache License 2.0.

What It Is

DVC is a data version control Git extension aimed at individual data scientists and small data science projects. Its stated goal is to bring software engineering practices to data, so datasets and ML work can be versioned and reproduced the way code is. The site presents it as a way to apply data version control to data science workflows with minimal overhead.

Where It Fits in the Stack

The DVC site splits its audience into two paths. For local workflows and small projects, DVC is the Git extension. For enterprise AI and data engineering teams, the site points to lakeFS, which it describes as highly scalable data version control infrastructure for petabyte-scale multimodal object stores and data lakes. A DVC extension for VS Code is also offered.

Community and Learning Resources

The project provides documentation, a getting-started guide, a blog, a course, a Discord chat, a forum and a GitHub issue tracker for bugs and feature requests. The site also lists a webinar on end-to-end lineage with Amazon SageMaker AI, MLflow and DVC.