Agentic AI Analytics Built for Data Teams

What this post added

This post introduces agentic AI analytics capabilities to Hex, enhancing the notebook experience with AI-powered exploration, generative app building, and code creation/editing. It also expands conversational self-serve analytics, allowing users to ask questions in natural language and receive governed insights. The Hex Agent is described as a pair programmer that understands data work, dependencies, and debugging. Threads are integrated into conversational workflows across platforms like Slack, Claude, Cursor, and Codex, bringing full stack context. Data apps can be turned into conversational interfaces where users can ask follow-up questions. The platform emphasizes accuracy and privacy, with secure metadata storage and no training on customer data by model providers. Governance is enhanced through context from semantic models, database metadata, and existing dashboards. Integration with external tools like Cursor and Claude Code is supported via a CLI.