Back office ops · Production

Da2a: An Open-Source Multi-Agent Data Platform Prototype Using the A2A Collaboration Protocol

The problem

Traditional data platforms required deep technical expertise and specialized data engineering, creating bottlenecks that left business decision-makers waiting days or weeks for answers and entirely dependent on overloaded engineering teams.

First attempt

The traditional approach required filing tickets with data engineering teams, building ETL pipelines, writing SQL queries, and waiting for reports — a slow sequential process where answers could become irrelevant before they arrived.

Workflow diagram · grounded in source
1
User asks orchestrator
trigger
“you simply ask a root "Orchestrator Agent."”
2
Orchestrator plans and delegates
ai_action
“The orchestrator understands the goal, formulates a plan, and collaborates with the specialist agents to get the answer”
3
A2A inter-agent communication
integration
“A2A provides a standardized way for agents to communicate their capabilities and call upon each other's skills over a network”
4
Domain agents execute sub-tasks
ai_action
“first asking the marketing agent to identify the sellers from the 'Display' channel, then passing that list to the e-commerce agent to calculate their total sales”
Reported outcome

The Da2a prototype demonstrates that a multi-agent system using the A2A protocol can answer complex cross-domain business questions by delegating sub-tasks to specialized agents, abstracting the underlying engineering complexity.

Reported metrics
Complex multi-step task handlingtackle complex, multi-step tasks that require synthesizing information from different domains
Engineering complexity abstractionabstracts the underlying engineering complexity
Reported stack
Da2aAgent-to-Agent (A2A) protocolAgent Development Kit
Source
https://mlops.community/blog/da2a-the-future-of-data-platforms-is-agentic-distributed-and-collaborative
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

The Da2a prototype demonstrates that a multi-agent system using the A2A protocol can answer complex cross-domain business questions by delegating sub-tasks to specialized agents, abstracting the underlying engineering…

What tools did this team use?

Da2a, Agent-to-Agent (A2A) protocol, Agent Development Kit.

What results were reported?

Complex multi-step task handling: tackle complex, multi-step tasks that require synthesizing information from different domains; Engineering complexity abstraction: abstracts the underlying engineering complexity (source-reported, not independently verified).

What failed first in this deployment?

The traditional approach required filing tickets with data engineering teams, building ETL pipelines, writing SQL queries, and waiting for reports — a slow sequential process where answers could become irrelevant befo…

How is this back office ops AI workflow structured?

User asks orchestrator → Orchestrator plans and delegates → A2A inter-agent communication → Domain agents execute sub-tasks.