Kaggle AI Days 2025: 5-Day AI Agents Intensive Course
Event Dates: November 10-14, 2025 Format: Online, Global Official Website: https://rsvp.withgoogle.com/events/google-ai-agents-intensive_2025 Kaggle Learn Guide: https://www.kaggle.com/learn-guide/5-day-agents
My Role and Contribution
Technical Leadership
- Position: Course Lead for Day 3: Context Engineering & Memory
- Name: (Sampath Kumar](https://www.kaggle.com/maddula)
- Title: Developer Programs Engineer, Google Cloud
- Responsibility: Led the curriculum development and implementation for context management in AI agents
Deliverables Created
- Colab Notebook 1: Short-term Memory Implementation
- Demonstrated session management techniques
- Practical examples using Google ADK SessionManagers
-
https://www.kaggle.com/code/kaggle5daysofai/day-5b-agent-deployment
-
Colab Notebook 2: Long-term Memory Implementation
- Showcased persistent memory patterns
- Integration with vector databases
- Production-ready memory architectures
- https://www.kaggle.com/code/kaggle5daysofai/day-5a-agent2agent-communication
Course Overview
This flagship Google AI course attracted over 420,000 registrations globally, following the success of the previous year's Gen AI Intensive Course. The course focused on building autonomous AI agents that go beyond simple chatbots.
5-Day Curriculum Structure
- Day 1: Introduction to Agents & Agentic Architectures
- Day 2: Agent Tools & Interoperability with MCP
- Day 3: Context Engineering & Memory (My Leadership Area)
- Day 4: Quality, Logging & Evaluation
- Day 5: Prototype to Production
Day 3 Focus: Context Engineering & Memory
As the Day 3 lead, I covered: - Short-term Memory: Conversation context and session management - Long-term Memory: Persistent knowledge and learning mechanisms - Memory Architecture Patterns: Best practices for production systems - Google ADK Integration: Practical implementation using ADK's memory components
Impact and Reach
- Community Engagement: Active participation in Discord discussions
- Capstone Projects: Participants used Day 3 concepts in their final projects
Technical Stack
- Primary Framework: Google Agent Development Kit (ADK)
- Language: Python
- Tools: Colab notebooks, Gemini API, Vector databases
- Platforms: Kaggle, YouTube (live sessions), Discord (community)
Key Takeaways for Participants
Through my Day 3 curriculum, participants learned: 1. How to implement stateful AI agents using memory systems 2. Best practices for context management in production 3. Practical patterns for short-term and long-term memory 4. Integration techniques with Google ADK
This course represents Google's commitment to democratizing AI agent development, and I'm proud to have contributed to educating the next generation of AI developers.