Sunset Learning Turns 30! Discover our story, exclusives, and celebrations →
→
- COURSE
ClearML ML Engineer Certification (CLEARML-ME)
Price:
Duration:
Certification:
Exam:
Continuing Education Credits:
Learning Credits:
This class prepares students for the ClearML Engineer certification. ClearML is an open-source MLOps platform that enables teams to seamlessly track, orchestrate, and scale machine learning workloads across Kubernetes, cloud, and hybrid environments. By the conclusion of this hands-on training, you will return to work with the skills to deploy, secure, and operate a full ClearML environment — from experiment tracking to GPU-powered model serving.
Throughout the course, you will learn to use Helm, Kubernetes, and cloud-native tools to manage ClearML at scale. You’ll configure external data stores, automate agent scaling, integrate with Hugging Face and vLLM, and practice troubleshooting real-world ClearML incidents. The curriculum combines scenario-based labs with production-focused simulations.
Upcoming Class Dates and Times
All Sunset Learning courses are guaranteed to run
Course Outline and Details
Prerequisites
- Python – PCEP Certification or Equivalent Experience
- Familiarity with ML frameworks (e.g., PyTorch, TensorFlow) and basic DevOps (e.g., Docker, CI/CD)
- Basic Linux command-line skills
Target Audience
- ML Engineers
- Data Engineers
- DevOps Engineers
- AI Platform Specialists
Course Objectives
- Build automated ML pipelines with ClearML orchestration and CI/CD in ≤30 minutes.
- Scale training and inference using queues and GPU agents.
- Monitor models for drift, performance, and operational health.
- Integrate Data Scientist outputs (e.g., Sarah’s models) into production pipelines.
- Collaborate with Data Scientists (Sarah) and Developers (Joe) using ClearML projects.
Course Outline
Introduction to ClearML for ML Engineers
- What is MLOps? Role in Production Workflows
- ClearML Overview: The Server, The SDK, and The Agent
- Introduction to ML Pipelines and Automation
- Set Up Python Environment and Install ClearML Agent
- Configure and Run a Local ClearML Agent (Worker)
- Run a Baseline Pipeline Script
Dataset and Model Versioning
- Ensuring Reproducibility with Data and Model Versioning
- Detecting Data Drift with Integrations (Evidently/Deepchecks)
- Model Management and Formats
- Version a Dataset
- Query and Validate Model Artifacts
- Simulate Drift and Trigger Alerts
Collaboration and Governance
- Collaborating with Data Scientists and Developers
- Meet Sarah: The Data Scientist Persona (Handoff Context)
- ClearML Projects and Team Visibility
- Governance with Model Cards for Compliance
- Access a Shared Project from Sarah
- Apply Model Card Metadata for Production
- Troubleshoot Pipeline Integration Issues
Automated Pipelines & Orchestration
- Building End-to-End ML Pipelines with ClearML
- Orchestration: Managing Queues and Agents
- Handling Pipeline Failures and Retries
- Define and Run a Pipeline (Data → Train → Eval)
- Modify Pipeline to Retry on Failure
- Trigger Pipeline via ClearML API
Deployment and Operations
- ClearML Serving: Deployment and Canary Strategies
- Monitoring Models for Drift and Performance
- Integrating Monitoring Tools (e.g., Prometheus)
- Deploy Model to ClearML Serving
- Set Up Drift Alerts for a Deployed Model
- Manage Compute Queues (Docker/CPU optimization)
- Execute a Canary Rollback (Traffic Update)
Capstone: Deploy a Real-World AI Pipeline
- Capstone Overview: Inventory Forecasting Pipeline
- Simulating Handoff from Data Scientist (Sarah)
- Certification Prep: Scenarios and Best Practices
- Build and Schedule Retraining Pipeline
- Automate Response to Drift (Trigger Retraining)
- Full System Test: Ingest → Train → Deploy
- Practice Certification Exam Tasks