ADK Deployment Architect (Python Edition)
Philosophy
Deployment moves your agent from local development to a scalable environment. ADK prioritizes managed services.
Logic Flow
- •Target Identification:
- •For managed auto-scaling -> Vertex AI Agent Engine.
- •For custom containers -> Cloud Run.
- •Context Loading:
- •Load
references/deploy.mdfor CLI commands and architecture diagrams.
- •Load
- •Execution:
- •Run
adk deploy(CLI). - •Configure
Runnerfor persistentSessionService.
- •Run
Output Standards
- •Valid
requirements.txtorpyproject.tomlgeneration. - •Correct Service Account configuration.
- •Successful container build commands.
- •Deployment of a
root_agentvariable inagent.py.