Build AI Agents on GKE in the GKE Turns 10 Hackathon

GKE 10th Anniversary Hackathon Summary

Key Concepts:

• Google Kubernetes Engine (GKE)
• Microservices
• Agentic AI
• Google AI Models (e.g., Gemini)
• Bank of Anthos
• Online Boutique
• Agent Development Kit (ADK)
• Model Context Protocol (MCP)
• Agent-to-Agent (A2A) Protocol

1. Hackathon Overview and Challenge:

• The Google Kubernetes Engine (GKE) is celebrating its 10th anniversary with a hackathon.
• The core challenge is to modernize existing GKE applications by integrating cutting-edge agentic AI capabilities.
• Participants are tasked with enhancing pre-existing microservice applications (Bank of Anthos or Online Boutique) with AI agents.
• The key constraint is that participants should not modify the core application code directly. Instead, they should build new components that interact with the existing APIs.

2. Prizes and Recognition:

• The hackathon offers a total of $50,000 in monetary prizes.
• Other prizes include Google Cloud credits, mentorship opportunities, and social media shout-outs.
• Winners may be featured on the Kubernetes Podcast, in a video with the GKE team, or win a sponsored trip to KubeCon North America.

3. Project Details and Requirements:

• Participants must use Google Kubernetes Engine (GKE) and Google AI models (like Gemini) to create intelligent agents.
• The goal is to elevate the chosen microservice application with these agents.
• Examples of agentic AI integration include:
  • An AI chatbot for Online Boutique that can query inventory, provide personalized product recommendations, or check a user's financial balance via the Bank of Anthos API.
  • An agent that monitors microservice performance on GKE, suggests troubleshooting steps, or automates remediation.
  • Unique or unexpected AI integrations with the chosen microservices.
• While GKE and Google AI models are mandatory, participants are encouraged to explore Agent Development Kit (ADK), Model Context Protocol (MCP), and the agent-to-agent (A2A) protocol for enhanced functionality.

4. Getting Started and Resources:

• Participants need to build an application to showcase the AI agents.
• Google provides resources to help participants get started:
  • Microservice applications (Bank of Anthos and Online Boutique)
  • Example agents on GitHub
  • Documentation
  • Quickstarts
  • Tutorials
  • Webinars hosted by Google experts
• All resources are available in the video description and on the hackathon home page.

5. Submission Requirements:

• To submit a project, participants need to provide:
  • A hosted project URL
  • A text description of the project
  • An architecture diagram
  • A URL to the public code repository
  • A demo video

6. Tips for a Winning Submission:

• The project must be new.
• The code must be publicly available for judging.
• The project must extend at least one of the two microservice applications (Bank of Anthos or Online Boutique) with AI agentic capabilities.
• GKE and Google AI models must be used.
• The demo video must demonstrate the project working and highlight the use of GKE and AI agents.

7. Bonus Points:

• Participants can earn bonus points by:
  • Publishing a public blog post, video, or podcast about their project.
  • Posting on social media using the hashtags #GKEHackathon and #GKETurns10.

8. Important Dates and Details:

• Submission details and dates are available on the registration page (link in the video description).

9. Technical Terms Explained:

• GKE (Google Kubernetes Engine): A managed Kubernetes service that simplifies container deployment, management, and scaling.
• Microservices: An architectural approach where an application is structured as a collection of loosely coupled, independently deployable services.
• Agentic AI: AI systems that can perceive their environment, make decisions, and take actions to achieve specific goals.
• Bank of Anthos: A sample microservices application that simulates a banking system.
• Online Boutique: A sample microservices application that simulates an e-commerce store.
• ADK (Agent Development Kit): A set of tools and libraries to simplify the development of AI agents.
• MCP (Model Context Protocol): A protocol for sharing contextual information between AI models and agents.
• A2A (Agent-to-Agent) Protocol: A protocol for enabling communication and collaboration between multiple AI agents.

10. Conclusion:

The GKE 10th Anniversary Hackathon challenges developers to leverage the power of GKE and Google AI to create innovative and intelligent microservice applications. By integrating agentic AI capabilities into existing applications like Bank of Anthos and Online Boutique, participants can explore new possibilities in cloud-native development and potentially win valuable prizes and recognition. The hackathon encourages the use of tools like ADK, MCP, and A2A to build robust and collaborative AI agents, pushing the boundaries of what's possible with GKE and AI.