Automated enterprise AI agent platform that builds production-grade agents optimized for knowledge retrieval, document intelligence, and governed data access across the Databricks Lakehouse.
Databricks' enterprise platform for building production AI agents — integrates with Unity Catalog for governance, MLflow for evaluation, and Vector Search for retrieval to deliver knowledge assistants that operate securely within the Lakehouse ecosystem.
Databricks Mosaic AI Agent Framework is an enterprise-grade platform for building, evaluating, and deploying production AI agents directly on top of the Databricks Data Intelligence Platform. Designed for organizations that have already standardized on the Lakehouse architecture, the framework gives data and ML teams a unified path from raw documents and structured tables to governed, retrieval-augmented agents that can answer questions, summarize unstructured content, and act on enterprise data without leaving the Databricks security perimeter.
At its core, the framework combines four tightly integrated capabilities. First, Mosaic AI Vector Search provides a serverless, fully managed vector database that automatically syncs with Delta tables, so embeddings stay current as source data changes. Second, the Agent Framework SDK offers a code-first authoring experience built around MLflow, LangChain, and LlamaIndex, letting developers compose RAG pipelines, tool-calling agents, and multi-step reasoning chains using familiar Python patterns. Third, the Mosaic AI Agent Evaluation suite enables systematic quality measurement through LLM-as-judge metrics, human review apps, and offline test sets that catch regressions before deployment. Fourth, Model Serving and AI Gateway provide low-latency endpoints with rate limiting, payload logging, PII detection, and unified billing across both open-source models (Llama, Mistral, DBRX) and proprietary providers (OpenAI, Anthropic, Google).
What distinguishes Mosaic AI from horizontal agent frameworks is its native integration with Unity Catalog, which extends row-level security, column masking, and lineage tracking directly into agent responses. This means an agent automatically respects the same data permissions a user has when querying tables, and every retrieval, tool call, and model invocation is logged for audit and governance. The platform also includes Agent Bricks, a newer AutoML-style capability that automatically optimizes RAG pipelines, instructed retrievers, and document intelligence workflows based on a customer's evaluation data, reducing the manual tuning typically required to reach production quality.
Mosaic AI is best suited for enterprises whose data already lives in Databricks or that are willing to consolidate on the Lakehouse for both analytics and AI workloads. Pricing follows the consumption-based DBU model used across Databricks, with separate charges for vector search, model serving throughput, and agent evaluation runs. While the platform offers depth that pure-play agent frameworks cannot match, it carries the learning curve and operational footprint of the broader Databricks ecosystem.
Was this helpful?
Revolutionary approach that eliminates manual trial-and-error agent development through Instructed Retriever technology, which automatically learns optimal retrieval strategies for each domain and query pattern, improving relevance by 15–25% over standard vector-search RAG.
Pre-built agent architectures optimized for common enterprise scenarios: Information Extraction agents for structured data extraction from documents, Knowledge Assistants for Q&A over document corpora, SQL Agents for natural-language analytics, and custom agents for specialized workflows.
Deep integration with Unity Catalog that enables agents to understand enterprise context including table schemas, column descriptions, data lineage, and access policies — allowing agents to answer questions with full awareness of organizational data assets.
Access to leading AI models from OpenAI, Anthropic, Google, Meta, and open source through the AI Gateway, with intelligent routing, cost tracking, rate limiting, and guardrails applied consistently across all model providers.
Comprehensive platform for monitoring, tracing, and optimizing AI agents with integrated experiment tracking, automated evaluation datasets, LLM-as-a-judge scoring, and production quality dashboards for continuous improvement.
Advanced capability that automatically creates domain-specific synthetic data resembling production queries, enabling teams to build robust evaluation suites and stress-test agents before deployment without relying solely on manually curated test sets.
Varies by SKU and region
Custom (negotiated)
$400 in DBU credits / 14 days
Ready to get started with Databricks Mosaic AI Agent Framework?
View Pricing Options →Databricks Mosaic AI Agent Framework works with these platforms and services:
We believe in transparent reviews. Here's what Databricks Mosaic AI Agent Framework doesn't handle well:
Weekly insights on the latest AI tools, features, and trends delivered to your inbox.
Through 2025 and into 2026 Databricks has continued to expand Agent Bricks with automated optimization for instructed retrievers and document intelligence agents, reducing the manual tuning previously required for production RAG. Mosaic AI Agent Evaluation has matured with broader LLM-as-judge metric coverage and tighter integration with human review apps for SME labeling. The platform has deepened support for governed tool calling through Unity Catalog functions, enabling agents to invoke SQL, Python, and external APIs under the same permission model as data access. AI Gateway has added more granular guardrails (PII detection, content safety) and broader third-party model coverage. DBRX and Llama family models remain first-class options on Foundation Model APIs alongside expanded partner model access via the gateway.
No reviews yet. Be the first to share your experience!
Get started with Databricks Mosaic AI Agent Framework and see if it's the right fit for your needs.
Get Started →Take our 60-second quiz to get personalized tool recommendations
Find Your Perfect AI Stack →Explore 20 ready-to-deploy AI agent templates for sales, support, dev, research, and operations.
Browse Agent Templates →