Google has announced a significant expansion to its Vertex AI platform, unveiling an open-source framework for building AI agents called Agent Development Kit (ADK). The new tool, announced at the Google Cloud Next conference, enables developers to build AI agents in under 100 lines of Python code, with support for more languages expected later this year.

The ADK is built on the same framework that powers Google Agentspace and Google Customer Engagement Suite (CES) agents, and is designed to simplify the development of multi-agent systems. According to VP of Cloud AI Saurabh Tiwary, the kit allows developers to shape how agents think, reason, and operate within guardrails, and select which large language models (LLMs) they run in the backend. The ADK also provides bidirectional audio and video capabilities, enabling interaction with agents beyond text.

Industry analysts expect the ADK to accelerate the adoption of multi-agent systems in enterprises. Gartner vice president analyst Jim Hare believes that making the framework open source will free up Google's engineering resources, allowing the company to leverage a community of developers to enhance and maintain the code base. "At the same time, Google gains by showing the developer audience it wants to help enterprises benefit from using agentic AI using open software and tooling," Hare said.

Duncan Van Kouteren, an analyst at Nucleus Research, notes that the open-source framework will also create a natural path for enterprises to move to Google's paid cloud services. "Google anticipates that once developers build with their framework, they're more likely to use Google Cloud for deployment," he said. Other cloud rivals, including IBM and Microsoft, have similar ambitions with their own open-source agent frameworks.

In addition to the ADK, Google has also introduced Agent Garden, a collection of ready-to-use, pre-built agent patterns and components designed to accelerate the model development process. This strategy is already being followed by Salesforce and Microsoft, which have introduced ready-to-use agent templates and components in their Agentforce and Copilot offerings, respectively.

Another significant update to Vertex AI is the introduction of Agent Engine, a fully managed runtime engine that helps developers deploy agents in production and maintain control over their behavior. Agent Engine is designed to handle infrastructure management, rebuilding agents or agent systems when moving to production, and supporting short-term and long-term memory to enable agents to recall past conversations and user preferences.

Van Kouteren sees Agent Engine as a way to eliminate the overhead of managing the infrastructure to run agents in production. "It handles all the behind-the-scenes work such as security, scaling, and monitoring, so enterprise teams can focus on what their AI agents do rather than worrying about how to keep them running properly," he said. Google plans to expand Agent Engine with computer-use and code-execution capabilities, and a dedicated simulation environment for rigorous testing of agents with diverse user personas.

The updates to Vertex AI demonstrate Google's commitment to advancing the development and deployment of AI agents, and its strategy to attract more enterprises to its cloud platform. As the AI landscape continues to evolve, it will be interesting to see how Google's competitors respond to these moves, and how the open-source ADK framework contributes to the growth of the AI ecosystem.