Tower - build and run data pipelines for your AI

What is Tower

Tower (docs.tower.dev, “pip install tower”) is a data flow platform for pipelines and agents with optional data management in Iceberg lakehouses.

Tower Use Cases

Tower provides Pythonic serverless compute, orchestration and lakehouse data storage for your AI project:

• Collaborate on Python apps with your team (with local development, application sharing, production deployments, orchestration and scheduling)
• Provide your AI Agents with fresh and reliable data (acquire data from databases and web APIs and store them in a lakehouse on S3)
• Use it for Feature Engineering (transform raw data to produce features for inference)

Using Tower during AI Hackathon

Each hackathon team gets 2 months of free Tower usage, including 5 users and 10,000 free compute minutes every month.

To claim this benefit, one of your team members should register at Tower. Then, under “Account Settings > Teams” create a new Team and invite your team members to the Team.

Send the team name to [email protected] to claim your team benefit.

Learning Tower Basics

Launch your 1st app in 10 minutes - go through the quickstart.

Collaborate with your team on building and running apps - learn to work in teams!

Provide agents with data - take a look at this example of acquiring data from a web API and storing it in a S3-based lakehouse. You can then run an agent that answers stock price questions and maintains a cache of prices in an Iceberg lakehouse table. The agent uses the first app as a Tool.