Machine Learning Engineering — Production ML Systems
Build ML systems that actually ship: data pipelines, training infrastructure, experiment tracking, model packaging, inference serving, monitoring, and CI/CD for ML. Everything a senior MLE needs to take a model from notebook to production.
Course Lessons
ML System Design — Requirements, Constraints & Architecture
Production Data Pipelines for ML
Feature Stores — Offline, Online & Point-in-Time Correctness
Training Infrastructure — GPUs, Distributed Training & Optimisation
Experiment Tracking, Model Registry & Reproducibility
ML Model Testing — Beyond Accuracy
Model Packaging — ONNX, TorchScript & Containers
Inference Serving — Latency, Throughput & Scaling
Production Monitoring, Data Drift & Model Decay
CI/CD for ML — Automated Testing & Deployment Pipelines
A/B Testing & Shadow Deployment for ML Models
ML Platform Architecture — Building for Scale