Loading...

About Us

VAST Orbit is built by a small team at VAST who believe data science should happen where the data already lives — and that the right tools should make that feel effortless.

The story behind VAST Orbit

The vision behind VAST Orbit Data kept growing. The tools didn't keep up.

For years the tools of data science forced a trade-off. Python and pandas are fast and flexible but bound by memory; Spark scales but is heavy and slow to iterate on; SQL is powerful but never quite fits the way a data scientist thinks. Meanwhile the data spread across databases, lakes, and files, and reaching all of it meant building yet another pipeline.

The VAST AI OS removed the usual reason for all that movement by unifying transactional and analytical processing, all-flash performance at data-lake economics, and linear scale from gigabytes to exabytes. What was still missing was a way for data scientists to use that power with the Python they already knew. VAST Orbit exists to close exactly that gap.

The philosophy of in-database computing Bring the tools to the data, not the data to the tools.

Everything in VAST Orbit follows from one principle: computation should happen where the data lives. Operations execute inside VAST, data stays put, and only results travel back to Python — so a petabyte dataset feels like a megabyte one, and nothing sensitive leaks out along the way.

Just as important, the experience stays familiar. You work with patterns you already know from pandas and scikit-learn, the heavy lifting happens in the database, and the same code you wrote in a notebook is the code that runs in production. VAST Orbit doesn’t move data to your tools; it brings your tools to the data.

The technical foundation of VAST Orbit Proven analytics, re-engineered for the VAST query engine.

Under the hood, VAST Orbit translates everyday Python into operations that run in the database: data preparation, statistical profiling, hundreds of analytic and SQL functions, multi-source joins across every catalog, and machine-learning inference — all executed in VAST rather than in Python.

Rather than reinvent that analytics surface from scratch, VAST Orbit builds on years of proven, open-source engineering, re-engineered for the VAST DataBase. That heritage is why the library arrived broad and battle-tested, and why so much of it works the moment you connect.

The team

Badr Ouali

DB Solutions Engineer

https://www.linkedin.com/in/badr-ouali/
Fouad Teban

Field Engineering Lead

https://www.linkedin.com/in/fouadteban/
Christian Neundorf

Senior Systems Engineer

https://www.linkedin.com/in/christian-neundorf-552a6721/
Chris Snow

Product Manager

https://www.linkedin.com/in/csnowuk/
Kiran Kumar

Principal Systems Engineer

https://www.linkedin.com/in/kiranmavatoor/
Maciej Paliwoda

Senior Systems Engineer

https://www.linkedin.com/in/maciejpaliwoda/
Kuldeep Venati

Principal Data Platform SE

https://www.linkedin.com/in/venati/

Get involved

VAST Orbit is young and moving quickly, which is the best time to help shape it. Whether you want to ask a question, report something broken, or contribute code, here is where to start.

Join the conversation

The fastest way to get help or share what you are building is the VAST Slack. If you are stuck connecting to a cluster, unsure how to express an operation, or just curious what others are doing with VAST Orbit, ask there — the team reads it. Join us at vastsupport.slack.com.

Report an issue

Found a bug or hit a query that should work but doesn’t? Open an issue on GitHub with a small reproduction and it helps us fix it quickly. File issues at github.com/vast-data/vastorbit/issues.

Improve the documentation

Contributions are not only about code. If something here was unclear, a missing example would have saved you time, or you spotted a mistake, a documentation pull request is one of the most valuable things you can send — it helps the next person who arrives with the same question.

Open a pull request

Want to add a function, fix behavior, or extend a model? We welcome it. Have a look at the Contribution Guidelines to see how the project is organized and how to get a change reviewed and merged.

Open source, maintained by VAST

VAST Orbit is open source and is developed and maintained by VAST Data. It is not a separately supported product and carries no enterprise SLA — but it is actively maintained, and the quickest ways to get help are the community Slack and the issue tracker above.

Note

VAST Orbit brings Python data science to the VAST AI OS: prepare, explore, analyze, and build AI — all with in-database execution at any scale.