DAO-ai orchestrator directing a team of specialist agents — researcher, planner, coder, writer, executor

DAO: Declarative Agent Orchestration¶

Production-grade AI agents defined in YAML, powered by LangGraph, deployed on Databricks.

DAO is an infrastructure-as-code framework for building, deploying, and managing multi-agent AI systems. Instead of writing boilerplate Python to wire up agents, tools, and orchestration, you define everything declaratively in YAML configuration files.

# Define an agent in 10 lines of YAML
agents:
  product_expert:
    name: product_expert
    model: *claude_sonnet
    tools:
      - *ai_search_tool
      - *genie_tool
    prompt: |
      You are a product expert. Answer questions about inventory and pricing.

YAML First — configure everything with simple YAML Python Powered — built for developers, by developers Modular & Extensible — add your own agents, tools, and capabilities Observability Built-in — logs, traces, and metrics out of the box

Getting started¶

Why DAO? — what DAO is and how it compares to other platforms.
Architecture — how DAO works under the hood.
Key Capabilities — 20 features for production agents.

Reference¶

Configuration Reference — the complete YAML schema.
CLI Reference — the dao-ai command-line interface.
Python API — programmatic usage and customization.

Guides¶

Examples — ready-to-use example configurations.
MCP Server — expose a dao-ai agent as a single MCP tool.
A2A Protocol — Google Agent2Agent endpoints on every Apps deployment.
Background Agents — kickoff / poll / cancel for multi-minute runs.
Auditable Tool Invocations — tamper-evident approval receipts and audit-trail queries.

Companion projects¶

DAO has two companion repos that make it easier to learn and to author configs:

DAO AI Workshop — a self-paced, hands-on workshop that takes you from zero to a deployed, governed multi-agent system. Organized as L100 → L200 → L300 with lectures and lab notebooks covering tool use, NL-to-SQL with Genie, vector search, memory, prompts + guardrails, and orchestration — all defined in YAML and running as a Databricks App. Start here if you're new to DAO.
DAO AI Builder — a React web app that provides a graphical interface for creating and editing DAO configurations. Explore capabilities, learn the configuration structure with guided forms, and build agents without writing YAML by hand. It generates valid configurations that work seamlessly with this framework.