MLOps: Deploying Machine Learning & Deep Learning Models

A model that only lives in a notebook is not finished.

Level: Intermediate
Modules: 6
Lessons: 19
Total time: 5h 47m

MLOps: Deploying Machine Learning & Deep Learning Models

Tools you'll use

Python
Docker
FastAPI
Kubernetes
GitHub

After this track you can

Serve a model as an API other applications can call
Package a model and its dependencies in a container
Track experiments, data and model versions
Detect data drift and know when to retrain

Taking a model from a notebook to production: packaging it, serving it behind an API, running it in a container, tracking experiments and versions, and then the part that decides whether it stays useful — monitoring for drift. A model that is accurate today degrades quietly as the world changes; this track teaches you to see it before your users do.

Read a free lesson: Why Models Decay in Production

What you'll learn, module by module

6 modules · 19 lessons

Every module and lesson in this track. Lessons marked free are open to everyone; the rest open once you own the track.

Every lesson in this track, with summaries

01
Module 1ML Systems in Production
Syllabus
Why models degrade after launch, how batch and online inference differ, and what to monitor.
1 free lesson4 lessons1h 4m4 simulations
1. 1Why Models Decay in ProductionFree
2. 2Training-Serving Skew
3. 3Batch vs Online Inference
4. 4An ML System Design Checklist
02
Module 2Packaging and Serving Models
Syllabus
From notebook to service: saving models, an API with FastAPI, and reproducible Docker containers.
3 lessons48 min
1. 1Saving Models Reproducibly
2. 2A Model API with FastAPI
3. 3Containerising with Docker
03
Module 3Serving Deep Learning Models and LLMs
Syllabus
GPUs, batching, quantisation and formats like ONNX — making large models fast enough and cheap enough.
4 lessons1h 6m
1. 1Latency, Throughput and Batching
2. 2ONNX and Model Optimisation
3. 3Quantisation
4. 4Serving LLMs
04
Module 4ML Pipelines and CI/CD
Syllabus
Experiment tracking, model registries, and automated pipelines from raw data to a released model.
3 lessons47 min2 simulations
1. 1Experiment Tracking and the Model Registry
2. 2Training Pipelines
3. 3CI/CD for Machine Learning
05
Module 5Monitoring and Retraining
Syllabus
Drift detection, performance with delayed labels, retraining strategies, and safe rollouts.
3 lessons48 min4 simulations
1. 1Detecting Data Drift
2. 2Retraining Strategies
3. 3Shadow and Canary Deployments
06
Module 6MLOps Capstone
Syllabus
One model, from notebook to a monitored cloud service — every part of this track in one project.
2 lessons1h 14m1 simulation
1. 1Choosing Cloud Infrastructure
2. 2Capstone: Deploy and Monitor