Get in Touch
 Duration 14 hours

Course Outline

Foundations of MLOps on Kubernetes

  • Core principles of MLOps
  • Distinguishing MLOps from traditional DevOps
  • Primary challenges in managing the ML lifecycle

Containerizing ML Workloads

  • Packaging models and associated training code
  • Optimizing container images specifically for ML workloads
  • Managing dependencies to ensure reproducibility

CI/CD for Machine Learning

  • Structuring ML repositories to facilitate automation
  • Incorporating testing and validation stages into the pipeline
  • Automating triggers for retraining and model updates

GitOps for Model Deployment

  • Key principles and workflows of GitOps
  • Utilizing Argo CD for streamlined model deployment
  • Implementing version control for models and configurations

Pipeline Orchestration on Kubernetes

  • Constructing pipelines using Tekton
  • Managing complex, multi-step ML workflows
  • Optimizing scheduling and resource allocation

Monitoring, Logging, and Rollback Strategies

  • Tracking data drift and monitoring model performance
  • Integrating alerting systems and observability tools
  • Implementing rollback and failover mechanisms

Automated Retraining and Continuous Improvement

  • Designing effective feedback loops
  • Automating scheduled retraining processes
  • Integrating MLflow for tracking and experiment management

Advanced MLOps Architectures

  • Deployment models for multi-cluster and hybrid-cloud environments
  • Scaling teams through shared infrastructure
  • Addressing security and compliance requirements

Summary and Next Steps

Requirements

  • A solid understanding of Kubernetes fundamentals
  • Practical experience with machine learning workflows
  • Proficiency in Git-based development practices

Target Audience

  • ML engineers
  • DevOps engineers
  • ML platform teams

Number of participants


Price per participant

Testimonials (3)

Upcoming Courses

Related Categories