Amperity launches Chuck Data AI agent for Databricks users

Amperity has introduced Chuck Data, an AI agent designed for customer data engineering within Databricks environments.

Purpose-built for data engineering

Chuck Data is positioned as the first AI assistant specifically tailored for customer data engineering tasks. The agent applies Amperity's identity resolution models, which are based on experience with billions of data sets from over 400 enterprise brands, to streamline and optimise common engineering processes.

Chuck operates directly in the terminal and integrates with users' Databricks lakehouse environments. By leveraging native compute and large language model (LLM) endpoints, the agent facilitates workflows such as identity resolution, data profiling, compliance tagging, and personal data identification.

According to Amperity, the reliance on manual coding and rules-based systems for preparing customer data has remained a significant bottleneck, even as data infrastructure has advanced. Chuck is intended to address these issues by enabling natural language command "vibe coding", allowing data engineers to delegate tasks to the AI assistant in plain English.

Key features and workflow

The feature set of Chuck Data centres on efficiency and compliance. Users can access a natural language interface for a range of data engineering operations, including:

• Identity resolution using Amperity Stitch on Databricks compute infrastructure
• Personal identifiable information (PII) tagging and customer profiling, interacting with Unity Catalog
• Zero-copy architecture, ensuring that Chuck does not move or replicate customer data in the process
• Compliance support through streamlined and accurate PII tagging inside Databricks

Derek Slager, Co-Founder and Chief Technology Officer at Amperity, explained the motivation for building Chuck Data and the impact it is expected to have for engineers:

"Customer data engineering is full of repetitive, painful work, so we built Chuck to get rid of it. Chuck understands your data and helps you get stuff done faster, whether you're stitching identities or tagging PII. No orchestration, no UI gymnastics—it's just fast, contextual, and command-driven."

Chuck can be installed with a single command. Engineers are then able to execute natural language tasks that ordinarily require complex coding, reducing manual effort and expediting the creation of accurate customer profiles at scale.

Integration and identity resolution

A distinctive feature of Chuck Data is the ability to run Amperity's identity resolution algorithms directly in the user's Databricks environment, replicating the technology from Amperity's established enterprise customer data platform. During its research preview phase, customers can use these capabilities without charge for datasets up to 1 million records, with additional credits provided for larger datasets.

Paid options will unlock unlimited data processing, access to Amperity's stable ID algorithms, and the provision of enterprise support for organisations with more demanding requirements.

Live demonstrations and capabilities

Amperity will be offering live demonstrations of Chuck Data's capabilities including resolving large volumes of customer identities rapidly, generating detailed customer ID graphs, and using artificial intelligence to streamline data engineering workflows.

The Chuck Data platform's intended function is to assist organisations to prepare and manage their customer data more efficiently, reducing reliance on manual coding and helping engineers to create business-ready insights in less time.