Building Agents with Google ADK and Vertex AI

סקירה כללית

This course guides you through building intelligent agents using the Agent Development Kit (ADK) and Google Cloud technologies. You will start with tool definition and agent tool usage, then progress through structured outputs, state management, and memory systems (short and long-term).

The course covers secure API integration, database interaction via MCP, web search with grounding, and Retrieval Augmented Generation (RAG). You'll explore multi-agent architectures and implement observability through distributed tracing.

For the final project, you'll build Betty's Bird Boutique Customer Service Agent that answers bird- and store-related questions.

סילבוס

• Introduction to Building Agents
Get to know your course instructors, set up GCP resources, and get an overview of the course.
• Extending Agents with Tools
Extend AI agents beyond text with tool integrations, enabling reliable real-time actions and data access.
• Implementing Agent Tool Usage with ADK and Vertex AI Gemini
Learn to extend LLM agents with ADK and Gemini for tool usage—integrate and register custom functions, handle errors, and guide agents with effective prompts for real-world tasks.
• Structured Outputs
Discover structured outputs in AI: transform responses into actionable JSON for integration. Utilize schemas, parsers, and function calls to enhance reliability and automation in workflows.
• Implementing Structured Outputs with Vertex AI Gemini and Pydantic
Learn to generate structured outputs with Vertex AI Gemini and Pydantic, enabling reliable data extraction and downstream processing using defined schemas and robust error handling.
• Agent State Management
Explore agent state management with state machines. Learn how agents track user input, instructions, and tool use for complex workflows, ensuring adaptability and reliability.
• Implementing Agent State Management with ADK
Learn to manage agent state using ADK through demonstrations, hands-on exercises, and quizzes for effective agent development.
• Short-Term Agent Memory
Explore short-term memory in AI agents, enhancing coherence via state, ephemeral, and ephemeral memory strategies for efficient context retention in active sessions.
• Implementing Short-Term Agent Memory with ADK
Learn to implement and apply short-term memory in agents using ADK, with step-by-step demos, hands-on exercises, and solution walkthroughs.
• External Tools and APIs
Explore using external APIs for real-time data, dynamic actions, and authenticating agents. Discover MCP, a protocol standardizing AI’s tool interoperability and safety.
• Implementing API Integration with ADK and Google Cloud Secret Manager
Learn to integrate APIs securely using ADK and Google Cloud Secret Manager with hands-on demos and practical exercises.
• Web Search Agents
Equip agents to search web for real-time, unstructured info. Ground responses in evidence using APIs, handle noise, and avoid hallucination for credible answers.
• Implementing Web Search Agents with ADK and Grounding with Google Search
Build web search agents using ADK, integrate with Google Search for grounding, and apply practical skills through demos and hands-on exercises.
• Interacting with Databases
Equip agents to access and modify structured data by using SQL for interaction and vector databases for semantic tasks, ensuring seamless integration with private systems.
• Implementing Database Interaction with ADK and MCP Database Toolkit
Learn to implement and practice database interaction using ADK and MCP Database Toolkit through demos, hands-on exercises, and guided solutions.
• Agentic Retrieval Augmented Generation
Discover Agentic RAG: Enhance RAG by enabling reflection, query reformulation, and intelligent adaption for nuanced answers. Master retrieval, reasoning, and retry loops.
• Implementing Single-Agent RAG with ADK and Vertex AI Search
Learn to build a RAG agent on Google Cloud using ADK and Vertex AI Search for querying custom unstructured data, including setup, search integration, and grounded response generation.
• Long-Term Agent Memory
Explore long-term agent memory: understand semantic, episodic, and procedural memories. Learn storage strategies and best practices for personalized, coherent interactions.
• Implementing Long-Term Agent Memory with ADK MemoryService
Learn to implement long-term conversational memory in agents using ADK and Vertex AI Agent Engine, enabling context recall and continuity across sessions with persistent memory storage.
• Agent Evaluation
Agent Evaluation guides assessing an agent’s task completion, quality, tool use, and system metrics using response, step, or trajectory strategies to ensure reliable and efficient operations.
• Implementing Agent Observability with ADK, OpenTelemetry, and Google Cloud Tracing
Learn to implement agent observability by configuring ADK with OpenTelemetry and Google Cloud Tracing to monitor, trace, and analyze agent interactions and performance.
• Betty’s Bird Boutique Customer Service Agent
Build an AI agent for Betty’s Bird Boutique that answers bird and store questions using databases, files, and web info, avoids orders and off-topic queries, and is tested in a dev environment.

---

נלמד על ידי

Henrique Santana and Allen Firstenberg

---

נושאים

Programming