Zum Inhalt

Build Long-running AI agents that pause, resume, and never lose context with ADK

Most agent tutorials stop at a stateless chatbot: a conversational loop that loses all memory as soon as the container restarts. Real enterprise workflows don’t wrap up in a single API call.. HR onboarding spans two weeks. Invoice disputes remain unresolved for days as they await responses from vendors. Sales prospecting sequences span numerous touchpoints throughout a one-month period. These processes are dominated by „idle time“ – long pauses where an agent sits dormant, waiting for a human signature, a shipping confirmation, or an approval gate. A stateless chatbot wouldn’t last long in that scenario. This tutorial demonstrates how to build a reliable New Hire Onboarding Coordinator Agent using the Agent Development Kit (ADK) that can run smoothly for weeks at a time. The agent sends a welcome packet, pauses for days while the employee signs documents, delegates IT provisioning to a specialized sub-agent, waits again for hardware delivery, and finally sends a personalized day-one schedule – all without losing a single byte of context.. Along the way, you’ll learn three architectural shifts that separate production agents from demo chatbots:. Use structured, persistent memory schemas rather than dumping raw JSON into a vector database.

  Google Developers Blog