PreviewWorth evaluating· Databricks release notes (AWS)
Managed agent memory and sessions are now in Beta
Databricks added Beta managed agent memory and managed agent sessions, two fully managed Lakebase-backed stores for agent state that work with agents built on any framework. Memory persists across conversations and supports semantic search; sessions store conversation history for the agent to read at the start of a turn and append to while it runs.
Learners building agents now need to distinguish long-term memory from per-session conversation history, and treat both as Beta.