CareerOS
Sign inStart for free
Learning Paths

ML Engineering & MLOps

Take a model from a notebook to a service someone depends on: clean training code, experiment tracking, packaging, deployment and monitoring for drift.

Finish this and you can train, version, deploy and monitor a model as a reproducible system, not a notebook that only runs on your laptop.

Starts at
Intermediate
Topics
5
Projects
5
Of material
135 h
  • Machine learning
  • MLOps
  • Python
  • Docker
  • Monitoring
Level
Format
  1. 01

    Practical machine learning

    Beginner

    Models that work on real data: problem framing, baselines, features, validation and the mistakes that make offline scores lie.

    Build this

    A model that beats an honest baseline

    Pick a public tabular dataset and show, with a proper split, how much a model actually adds.

    1. Frame the problem and the metric a business would care about.
    2. Build a baseline with no ML at all, then a simple model, then a better one.
    3. Use a pipeline so preprocessing is fitted on training data only.
    4. Report the gain over the baseline with a confidence interval.

    Done when: A notebook and a one-page summary of the gain over the baseline.

    UCI Machine Learning Repository
  2. 02

    Production-grade ML code and experiments

    Intermediate

    From notebook to package: tests, configuration, data versioning and experiment tracking so every result can be reproduced.

    Build this

    Training anyone can reproduce with one command

    Turn the baseline project into a package.

    1. Move code out of the notebook into modules with unit tests for data validation and features.
    2. Drive runs from configuration, not edited code.
    3. Track every run with MLflow, including the data version.
    4. Prove it: a fresh clone and one command reproduce your best run.

    Done when: A repository where one command reproduces the best run, with tests passing in CI.

  3. 03

    Packaging and serving models

    Intermediate

    Containers, model APIs, batch versus online inference, and deploying to the cloud.

    Build this

    A model service with a health check and a load test

    Serve your reproducible model behind an API.

    1. Wrap it in a small web API with input validation and a health endpoint.
    2. Containerise it and run it locally with compose.
    3. Deploy it to any free cloud tier.
    4. Load-test it and report latency at p50 and p95.

    Done when: A live endpoint, the Dockerfile and the load-test results.

  4. 04

    Monitoring, drift and retraining

    Advanced

    How models fail quietly in production, how to notice, and when to retrain.

    Build this

    Catch drift before your users do

    Add monitoring to your model service.

    1. Log inputs and predictions from the live service.
    2. Simulate drift by shifting one feature's distribution in replayed traffic.
    3. Detect it with a drift report and an alert threshold you can defend.
    4. Write the retraining runbook: who acts, on what signal, and how you roll back.

    Done when: The drift report, the alert, and the runbook.

  5. 05

    ML systems design

    Advanced

    Design an ML system on a whiteboard: requirements, data, features, training, serving, evaluation and failure modes.

    Build this

    A design document for a recommender

    Write the document a team would review before building.

    1. State requirements: scale, latency, freshness and the business metric.
    2. Design data and features, including how labels are collected.
    3. Choose training and serving patterns, with the alternatives you rejected and why.
    4. List failure modes and how each would be detected.

    Done when: A four-to-six-page design document, reviewed by one other engineer.

Not sure this is your direction?

A three-minute counselling sample is free.

Try counselling