Databricks Acquires Quotient AI to Enhance Evaluation of AI Agents

Databricks has acquired Quotient AI, a company specializing in evaluation and reinforcement learning for AI agents. The acquisition aims to enhance continuous monitoring and improvement of AI agents across Databricks products, including Genie, Genie Code, and Agent Bricks.

Quotient AI was founded by engineers who previously worked on quality improvements for GitHub Copilot. Its platform analyzes complete agent traces to detect issues such as hallucinations, reasoning errors, and incorrect tool use. These insights are converted into structured datasets and reward signals that enable organizations to monitor and fine-tune AI agents in production environments.

By integrating Quotient’s technology, Databricks adds a continuous evaluation layer to its platform, allowing AI agents to improve their performance over time. This integration supports enterprise customers deploying AI systems in critical workflows, providing tools to measure and address performance issues at scale.

The acquisition also strengthens Databricks’ new product Genie Code, an autonomous AI agent designed to assist data teams in planning, building, and running data engineering, machine learning, and analytics workflows. Together with Genie and Agent Bricks, the combined capabilities aim to deliver more accurate and reliable AI systems for enterprise use.