Travis Burmaster

I lead the AI Architecture practice at Next Link Labs. Before that I spent the last decade building agentic AI systems, multi-agent platforms, and the unsexy plumbing that makes them actually run in production.

The Databricks practice exists because most enterprise AI initiatives fail in the same place: memory. Models forget. Agents lose state. RAG can't keep track of a conversation. And the workaround is always a parallel data system (a sidecar vector DB) with its own governance, its own access control, its own lineage. That's not a system you can ship.

So we built a practice around making memory a first-class citizen on the Databricks Lakehouse you already have. Claude shows up where it earns its keep. Everything else lives in Unity Catalog.