Databricks has released KARL, an RL-trained RAG agent that it says handles all six enterprise search categories at 33% lower cost than frontier models.

Databricks' KARL agent uses reinforcement learning to generalize across six enterprise search behaviors — the problem that breaks most RAG pipelines.

As data sources and volumes grow, and as a data-driven orientation is increasingly deemed to be a competitive necessity, the war between platform vendors to provide the primary repository for our data ...

Agent Bricks uses technology developed by MosaicML’s research division to optimize the agent development process and help control the cost of running agents. Databricks has released a beta version of ...