Seamless Data Coordination

Dataset version control and coordination for ML teams.

Stack is as powerful as Git yet as simple as Google Sheets. 

Thanks for joining 🚀

Stack Dataset Homepage - Images.jpg
 

Before Stack

WhatsApp Image 2022-08-04 at 6.50.51 PM.jpeg

After Stack

After Stack IMG.png

Problem

ML teams are consistently analyzing the root cause of dataset issues.

The analysis involves searching over old snapshots and backups of the dataset. 

Teams resort to e-mail and slack to communicate changes to the data.

This process costs data teams thousands of work hours every year.

Solution

Stack is a simple approach to version control and coordinate datasets.

Our tool connects to cloud storage, versioning every change in the dataset.

We track the author, date, and description of each change, along with a “diff”.

Stack seamlessly gives you the story of your data.

ezgifcom-gif-maker-1.gif

Product

Command-line and web interfaces.

Connects to your storage and automatically tracks interaction with your data.

Generated customizable data alerts for your team via e-mail.

Indexes the history of the dataset, compares different versions and creates endpoints to stream data for model training.

Agnostic to data source or schema and integrated with common cloud storage.

Stack Dataset - Activity health care use
Stack Dataset Homepage - Table (1)_edite
Stack Competition Analysis (6).jpg

Our team

Attachment_1633105570 (1).jpeg

Bernardo Aceituno
CEO.

MIT Ph.D. CS,

ex-Facebook and ex-founder.

  • LinkedIn
Attachment_1633105570 (2).jpeg

Melissa McAneny
COO.

MIT MBA,

ex-SpaceX and ex-Tesla.

  • LinkedIn
Attachment_1633105570.jpeg

Antoni Rosinol
CTO.

MIT Ph.D. CS,

ex-NASA, ex-GoPro, and ex-founder.

  • LinkedIn