Databricks Inc. introduced Genie Code, an artificial intelligence agent built to automate complex data engineering and analytics tasks, while also acquiring Quotient Artificial Intelligence Inc., a startup that evaluates and diagnoses failures in artificial intelligence agents. Genie Code is positioned to move data teams from basic code completion tools toward systems that can autonomously plan and execute data workflows under human supervision. Databricks executives framed the launch as part of a broader shift in enterprise data work, where the goal is not just to generate code but to understand and act on the data context embedded in corporate systems.
Genie Code is designed to integrate deeply with enterprise data systems and governance layers so it can interpret organizational data context, historical query patterns and business definitions, then translate user intent into definitions needed for production workflows. Databricks’ Unity Catalog provides the governance and security boundary, and Genie Code is intended to run primarily on the Databricks platform, with the option to connect external sources via Unity Catalog. Company leaders said agents are changing the role of data professionals by shifting work away from writing code toward supervising and orchestrating artificial intelligence agents, with significant productivity gains expected not only in development but in the operational maintenance of data systems such as keeping pipelines running and troubleshooting upstream changes.
Databricks technologists reported that Genie Code is already automating labor-intensive preparation tasks such as cleaning tables, finding missing values, imputing them and performing transformations, freeing data scientists to focus on core machine learning. The acquisition of Quotient Artificial Intelligence is intended to improve reliability and performance of these agent-based systems using reinforcement learning models that analyze agent behavior and identify where processes break down. Quotient Artificial Intelligence, founded by developers behind GitHub Inc.’s Copilot, trains custom models that can review an agent’s activity and determine when it made the wrong tool call. Databricks plans to integrate Quotient’s technology into Genie Code and into its broader agent platform so organizations can continuously monitor deployed agents, understand their mistakes and adapt to changing environments over time.
