This course teaches you to design, implement, and deploy multi-agent systems using ADK and Vertex AI Gemini. You'll master 8 core agentic workflow patterns through hands-on exercises and apply them in a comprehensive capstone project.
- Python programming experience
- Basic understanding of LLMs and APIs
- Google Cloud Platform familiarity
| Module | Focus | Duration | Key Skills |
|---|---|---|---|
| Lesson 1 | Workflow Analysis & ADK Basics | 60 min | Task decomposition, agent coordination |
| Lesson 2 | Code-First Workflow Modeling | 45 min | Agent classes, data flow design |
| Lesson 3 | ADK Implementation Patterns | 60 min | SequentialAgent, ParallelAgent, Gemini integration |
| Lesson 4 | Prompt Chaining Pattern | 45 min | Sequential validation, context passing |
| Lesson 5 | Routing Pattern | 45 min | Input classification, intelligent routing |
| Lesson 6 | Parallelization Pattern | 45 min | Concurrent execution, result aggregation |
| Lesson 7 | Evaluator-Optimizer Pattern | 60 min | Iterative refinement, quality loops |
| Lesson 8 | Orchestrator-Workers Pattern | 60 min | Dynamic delegation, multi-agent systems |
AI Research Assistant - Build a comprehensive research agent that demonstrates all 8 workflow patterns in an integrated system for automated knowledge synthesis.
# Clone the course repository
git clone [repository-url]
cd course-2-agentic-workflows-adk
# Install dependencies
pip install -r requirements.txt
# Set up GCP authentication
gcloud auth application-default login
gcloud config set project YOUR_PROJECT_ID
# Enable required APIs
gcloud services enable aiplatform.googleapis.com# Test ADK availability
python -c "from google.adk.agents import LlmAgent; print('ADK ready!')"
# Test Vertex AI access
python -c "from google import genai; print('Vertex AI ready!')"