Overview

Introduction

We are giving you AI teammates to help with data work ranging from building data pipelines to managing your data infrastructure. We call these teammates "DataMates", and they can work with your entire tool stack like Databricks, Snowflake, Airflow, dbt, Jira, Cursor, GitHub Copilot, Windsurf, and many others.

Here are some example work items that datamates can do -

1. Translate data pipeline from t-sql to snowflake SQL, provide explanations and migrate underlying tables
2. Generate Airflow dag based on templates created in Confluence and some previous examples
3. Read a bug filed in JIRA for a dbt model and suggest code changes, and create PR in Git
4. Analyze Snowflake SQL for performance by analyzing query profile, underlying table partitions, and create PR in Git with optimized query code

Plus, of course, there are many more. Check them out here!

Components

The DataMates platform comes with following components -

1. Knowledge Hub You can link your best practices, cookbooks, and other knowledge in the form of Google docs, Confluence pages, and other documents
2. Memory Hub AI teammates remember your work patterns from your coding assistants directly as "memory" so they work exactly like you would
3. Tool Integrations All different tool integrations like Databricks, Snowflake, Airflow, dbt, JIRA to pull the necessary context and execute tasks in these different tool systems. Also, the ability to integrate your own tools via direct APIs.
4. Datamates (AI Teammates) Each AI teammate is specialized at certain tasks. The platform comes with a few datamates out of the box, and you can also create some of your own.

Here's a quick overview Demo of one of tasks that Datamates can do -