Some of your company’s most valuable data is still hard to access. Documents, slides, PDFs, and internal systems contain important information, yet teams spend too much time finding, cross-checking, and piecing together references rather than making decisions.
Traditional approaches such as Retrieval Augmented Generation (RAG) involved complex parsing, chunking, and embedding of unstructured data and constant updating and optimization to maintain while providing measurable results. This made it impossible to access many teams and systems, locking your data.
Knowledge Assistant changes this. Now generally available and expanded to 10 new fields, you can turn your documents into accurate, on-the-ground answers in just minutes. As part of the Agent Bricks platform, it provides a fully managed experience across the entire agent lifecycle, from ingestion and continuous updates to retrieval and inference, with a scalable endpoint that you can integrate anywhere. Knowledge Assistant is powered by Databricks AI research and achieves 70% more Reply quality compared to the simplified RAG approach, without the operational overhead.
A new approach to the recovery of real enterprise data
Most retrieval systems were designed for a simple world: a single index, a uniform schema, and similarity search as the primary hint. Enterprise knowledge doesn’t look like that. It spans multiple systems, each of which has different structures, metadata conventions, and expectations regarding freshness and authority. Treating them all as the same data source limits both quality and control.
Knowledge Assistant is built on a fundamentally different architecture, developed by the Databricks AI research team instructed the retriever. Rather than relying solely on similarity search, the assistant understands how each knowledge source is organized and how it should be interrogated. When a user asks a question, the Knowledge Assistant translates that request into precise, source-aware questions – including guidance like prioritizing recent content, emphasizing specific metadata, or favoring authoritative sources.
This approach allows the assistant to synthesize information across different systems while remaining grounded in the structure of each system. it means you got it page-level citations on each response, Reducing hallucinations and allowing your users to quickly get back to the source.
Because this architecture is delivered as a fully managed service, teams benefit from ongoing research improvements automatically, rather than preventing recovery quality at the time of an agent deployment. We are constantly evaluating new models, techniques and research, running it against our extensive evaluation suite, and then automatically incorporating it into your agent in a seamless way.
Bridging the gap between subject matter experts and developers
Most enterprise AI systems rely on static signals, labels, or coarse feedback signals that capture individual answers but fail to generalize. Knowledge Assistants are used differently Agent learning from human feedback (ALHF) to translate expert guidance into sustainable, repeatable improvements in agent behavior.
Unlike one-off improvements, ALHF generalizes expert feedback throughout the conversation. Your experts only need to provide questions and guidelines for answers, and Knowledge Assistant will do the rest. Every response is captured in an end-to-end trace and logged as structured, governed data. With native MLflow integration, teams can evaluate changes and track quality improvements with the same rigor used for production ML systems.
“Agent Brix turns our code into conversations, an always-available AI teammate that helps us move faster, work smarter, and focus on providing frictionless revenue management.” -Ben Bartholik, Principal Data Engineer, FinThrive
Create your Knowledge Assistant Agent today
Getting started is simple. Upload your documents and Knowledge Assistant takes care of the rest. Use Knowledge Assistants to create agents who understand market research, support documentation, or policies and procedures to unlock your enterprise knowledge and accelerate productivity.
