Google ADK - Bootcamp

Duration: 1 Month (12 Lectures – 3 per week)
Mode: Hands-on, Instructor-Led
Level: Beginner to Advanced

Mode: Online

Fee: Rs. 5,000/-

Course Overview

This intensive crash course is designed to take you from the fundamentals of AI agents to building advanced, autonomous, and multi-agent systems using Google’s Agent Development Kit (ADK). With a strong balance of theory and hands-on coding, you’ll learn how to design, build, and deploy intelligent agents for real-world applications.

Course Schedule

Week 1 – Foundations of Agent Development

Lecture 1 – Introduction to AI Agents Setup

What are AI Agents? Core principles of perception, reasoning, and action

Introduction to the Google ADK (architecture & workflow)

Environment setup and first agent demonstration

Lecture 2 – Building Your First Agent

Understanding the agent lifecycle

Designing inputs, prompts, and outputs

Extending a simple agent to handle multiple queries

Lecture 3 – Tool-Enhanced Agents

Why agents need tools and external integrations

Adding tools to extend agent capabilities

Practical example: connecting an agent to a custom tool (e.g., weather, calculator)

Week 2 – Enhancing Agent Intelligence

Lecture 4 – Integrating Multiple LLMs

Why use LiteLLM for flexibility

Switching between different LLM providers (OpenAI, Anthropic, etc.)

Comparing performance and accuracy across models

Lecture 5 – Structured Outputs

The importance of structured responses in production

Enforcing JSON and schema-based outputs

Hands-on example: agents returning validated data

Lecture 6 – Sessions & State Management

How agents maintain context across conversations

Short-term vs. long-term memory design

Building agents that adapt to user preferences

Week 3 – Memory Collaboration

Lecture 7 – Persistent Storage

Persisting conversations with databases or files

Trade-offs: performance vs. memory size

Real-world example: storing and retrieving chat history

Lecture 8 – Multi-Agent Systems

Why multi-agent collaboration matters

Designing specialized agents that work together

Example: researcher agent + summarizer agent collaboration

Lecture 9 – Stateful Multi-Agent Systems

Combining state management with multi-agent collaboration

Handling complex workflows across multiple agents

Example: a travel booking system with planner, budget, and confirmation agents

Week 4 – Advanced Workflows, Capstone

Lecture 10 – Callbacks Monitoring

Tracking the lifecycle of agent actions

Debugging and monitoring with callbacks

Implementing logging and usage metrics

Lecture 11 – Sequential, Parallel Workflows

Sequential task pipelines vs. parallel task execution

When to use each for efficiency and accuracy

Real-world scenarios where hybrid workflows shine

Lecture 12 – Autonomous Agents, Capstone Project

Autonomous agent loops

Safety considerations and guardrails for autonomous systems

Capstone Project: Build a fully functional autonomous agent system

Features: persistent memory, structured outputs, tool integration

Example use case: a news summarizer with fact-checking and storage

Learning Outcomes

By the end of this course, you will be able to:
✔️ Understand AI agent fundamentals and ADK architecture
✔️ Build and extend agents with tools, memory, and persistence
✔️ Design multi-agent collaborative systems
✔️ Implement monitoring, debugging, and optimization strategies
✔️ Create safe and autonomous agents for real-world applications