GKE AI Labs

Unlock Scalable and Efficient AI Workloads with Google Kubernetes Engine (GKE)

Experiment with the latest tools and practices for deploying and scaling AI on GKE

Discover practical examples of AI Agents deployments and scaling using GKE

Quickly launch pre-configured solutions for modern AI models inside Google Kubernetes Engine infrastructure

Follow structured tutorials to master GKE key concepts and skills

GKE AI/ML Use Cases

Explore how AI on GKE empowers the entire AI/ML lifecycle, from initial experimentation to scalable production deployments. Learn how Google Kubernetes Engine (GKE) provides robust tools and infrastructure to streamline AI workloads, enabling faster experimentation, efficient orchestration, and reliable model serving. Discover practical examples and best practices for leveraging AI on GKE to accelerate your AI/ML initiatives and achieve scalable, production-ready solutions.

Experimentation

Leverage AI on GKE to manage containerized environments, simplify dependencies, and ensure reproducibility. Explore practical examples of using Google Kubernetes Engine (GKE) for distributed training and hyperparameter tuning, enabling faster experimentation cycles and accelerating your AI/ML workflows.

Creating Inference Checkpoints Fine-tuning Gemma 3-1B-it on L4
Orchestration

Streamline your AI on GKE workflows with Google Kubernetes Engine's powerful orchestration capabilities. Manage complex pipelines, schedule jobs, and automate resource allocation to optimize your AI/ML initiatives. Leverage AI on GKE to simplify orchestration and achieve scalable, production-ready solutions for your machine learning models.

Workflow orchestration GKE cross region capacity chasing with SkyPilot DWS
Evaluation

Ensure the quality and performance of your AI on GKE models with Google Kubernetes Engine's robust evaluation infrastructure. Deploy evaluation services and dashboards to monitor key metrics, track model performance, and gain actionable insights. Leverage AI on GKE to optimize your AI/ML workflows and achieve reliable, production-ready solutions.

Deploying a containerized agent built with the Google Agent Development Kit (ADK) that uses VertexAI API Building Agents with Agent Development Kit (ADK) on GKE Autopilot cluster using Self-Hosted LLM Building Agents with Agent Development Kit (ADK) on GKE using Ray Serve for Self-Hosted LLMs
Serving

Deploy and scale your trained AI on GKE models with Google Kubernetes Engine's production-ready serving infrastructure. Leverage containerized serving solutions to deliver low-latency predictions, handle high traffic volumes, and ensure reliable performance. Optimize your AI/ML workflows with AI on GKE to achieve scalable, efficient, and production-ready solutions.

Kserve Deploying a Persistent Chatbot on Google Cloud Platform with LangChain, Streamlit, and IAP Llamaindex in GKE cluster

AI on GKE Community & feedback

Explore a curated selection of impactful AI on GKE projects and groundbreaking research from our community. These innovative contributions showcase the transformative potential of AI/ML technologies and are shaping the future of artificial intelligence. Discover how AI on GKE is driving advancements in distributed training, scalable model serving, and efficient orchestration. Join our community to share your expertise, collaborate on cutting-edge projects, and help us improve our learning materials. Together, we can accelerate innovation and create scalable, production-ready AI solutions with AI on GKE.

Share your knowledge and help us improve our learning materials.