ML Engineering and MLOps.
Build your next skill.

Format: 35 training days, live online. Foundation-to-project coverage with guided practice. Exact pace and batch schedule are agreed with the trainer.

Prerequisites: Working Python, pandas, basic statistics and introductory machine learning.

Days 1-5 | ML and software foundations

• Python environments, Git, testing and project structure
• Data splits, leakage, baseline models and metric selection
• Train and evaluate a small scikit-learn model

Days 6-10 | Reproducible experiments

• Feature preprocessing, validation and reproducible runs
• MLflow experiment tracking, parameters, metrics and artifacts
• Data/model versioning and model registry concepts

Days 11-15 | Reliable training pipelines

• Modular ingestion, preprocessing, training and evaluation steps
• Pipeline orchestration, configuration and retry behaviour
• Automated validation, model comparison and promotion criteria

Days 16-20 | Model serving

• Batch inference vs online APIs and model packaging
• Docker, API contracts, validation and inference tests
• Latency, throughput, rollout and rollback fundamentals

Days 21-25 | CI/CD for ML

• Automated linting, unit tests and data/model checks
• Build/deploy workflows with environment separation
• Model approval gates and infrastructure/secrets basics

Days 26-30 | Monitoring and governance

• Service health, feature drift and prediction-quality monitoring
• Retraining triggers, delayed labels and human review
• Access controls, audit trail, model cards and cost awareness

Days 31-35 | Capstone and interview practice

• Ship a tracked, tested model pipeline and small inference service
• Demonstrate monitoring checks, a new model version and rollback
• Explain trade-offs, present the project and practise interviews

Training and interview preparation

• Live online training with guided exercises and project discussion.
• Interview preparation, technical questions and mock interview practice.
• Live demo video available. Ask the team to arrange the demo and confirm the batch details.

This is a practical learning plan, not a promise of certification, a job or production-level expertise in 35 days. Cloud labs or model APIs may have usage costs; confirm lab access with the trainer before starting.