Ivan Leo explains Google’s Interactions APIAn application programming interface is a defined way for software systems to request data or actions from one another. as an interface for model and agent workloads. He describes preserving conversation contextAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts. through interaction identifiers and handling typed multimodal outputs and tool callsAI tool calling lets a model request a structured operation from software instead of only returning natural-language text..
He demonstrates a managed coding agentAn AI coding agent is a tool-using AI system that can inspect, modify, and validate software within a repository. using a remote sandboxAn AI agent sandbox is an isolated execution environment that limits which files, processes, networks, credentials, and external systems an agent can access., retaining workspace state across calls and accepting project sources. A repository analysis example illustrates a longer agent task.
Leo describes a proxy that injects credentials into selected requests without exposing the secret to the agent, and discusses configured agents and migration tools. The capabilities and service claims are presented as Google’s account.
Watch on YouTube
