Vertex AI unveils tools for multi-agent ecosystem development

Vertex AI, a comprehensive platform for creating and managing AI agents, has introduced new capabilities aimed at enhancing the capacity of enterprises to build and deploy multi-agent ecosystems.

Saurabh Tiwary, Vice President of Cloud AI, has stated that "every enterprise will soon rely on multi-agent systems – multiple AI agents working together – even when built on different frameworks or providers. Agents are intelligent systems that can act on your behalf using reasoning, planning, and memory capabilities. Under your supervision, they can think multiple steps ahead and accomplish tasks across various systems."

The platform now includes tools to aid development and connectivity of AI agents, leveraging models such as Gemini 2.5, allowing seamless integration with enterprise workflows and data. The introduction of the Agent Development Kit (ADK) assists developers in building AI agents with just under 100 lines of Python code. The ADK supports various models, providing a flexible choice for developers, as well as supporting Model Context Protocol (MCP) for secure data connections.

Aakriti Bhargava, Vice President of Product Engineering and AI at Revionics, remarked, "Using Agent Development Kit, Revionics is building a multi-agent system to help retailers set prices based on their business logic — such as staying competitive while maintaining margins — and accurately forecasting the impact of price changes. ADK streamlines multi-agent transfer and planning, such as knowing when to transfer between specialized agents (data retrieval) and tools (constraint application), thereby combining Revionics' pricing AI with agentic AI to automate entire pricing workflows. Data is central to Revionics' process, and the development kit enables agents to efficiently reason over big data through storage artifacts rather than relying solely on the LLM context."

The "Agent Garden" component offers developers ready-to-use samples and tools, facilitating development with pre-built agent patterns and components. ADK further integrates with Vertex AI for a frictionless transition from development to production deployment. The ADK is optimised for both Gemini and Vertex AI, featuring enhanced reasoning capabilities that make it possible to deploy agents to a managed runtime enterprise-wide.

Laurent Giraud, Chief Data (and AI) Officer at Renault Group, noted their use of the ADK, "We used the ADK to develop an agent that ensures we're installing EV chargers where drivers need them most. The agent assists our data analysts to leverage geographical, zoning, and traffic data to inform and prioritize critical EV infrastructure investments that maximize driver convenience with less strain on our teams."

Moreover, the fully-managed "Agent Engine" provides a global, secure scale for deploying custom agents, supported by built-in testing and evaluation tools. "We've implemented the Agent Engine as the backbone of our video analysis AI agent, powered by Gemini," said Rina Tsuji, Senior Manager of Corporate Strategy at Nippon Television Holdings. "This setup allows us to leverage the Python Vertex AI SDK without worrying about infrastructure, saving us an estimated month of development time. Plus, the Agent Engine's API seamlessly connects with other Google Cloud products like Workflows, giving us excellent maintainability and room to grow."

Alongside ADK and Agent Engine, the "Agent2Agent" protocol offers a common language for interoperability among agents regardless of framework or vendor. This initiative currently collaborates with over 50 industry leaders including Box, Deloitte, and Salesforce to advance multi-agent system capabilities.

The Vertex AI platform also provides multiple security measures, including controlling agent outputs with Gemini's built-in safety features and managing agent permissions to prevent data breach risks. Enterprises can confine agent activity within secure perimeters using Vertex AI's security controls, along with tracing capabilities to monitor agent behaviour.