AI Engineering / 6 weeks
MLOps and AI Platform
Teaches engineers and platform leaders how to ship, monitor, and operate models reliably across training, deployment, evaluation, and incident response.
Learning path
Six modules with practical application.
- Versioned datasets, code, configs, and models flowing through training, evaluation, deployment, and retirement
- Reproducible training pipelines and a single source of truth for every experiment run, metric, and hyperparameter
- Online, batch, edge, and serverless deployments, plus shadow, canary, and champion challenger patterns to ship safely
- Real time monitors for performance, drift, calibration, latency, cost, and safety with thresholds that actually fire
- Held out eval sets, online experiments, and human feedback that keep telling you whether the model is still good enough
- Detecting, communicating, mitigating, and learning from AI incidents with the same rigor as any production incident
Loading course material…