Home › Paths › ML Engineering › Open source
Learning path · Open source
ML Engineering on Open source
You turn models into reliable products: features, training, serving and monitoring.
Free, like everything here: no sign-up, no paywall. Only the cloud you practise on may charge; the open-source path costs nothing.
Classic machine learning still runs most predictions in business: churn, fraud, demand, pricing. ML engineers make those models repeatable, served, monitored and retrained, which is also the foundation for LLM work.
Every topic below is a concept (the job) with the Open source product that does it. Learn the concept once; on another platform only the product name changes. Read alongside: The AI Universe · Certification prep on this site.
Beginner · 8 weeks
Crawl: You can build one thing end to end, with help.
Train a model honestly, track it and serve it.
Key ideas at this level, the same on every platform:
Learn
| Learn | What it means | On Open source you use |
|---|---|---|
| Python for data | pandas or Polars, plotting, notebooks | Jupyternotebook workspace |
| Core ML | train, validate, test; leakage; a dumb baseline first | scikit-learn, PyTorchtraining |
| Track experiments | every run, its data and its score | MLflowexperiment tracking |
| Serve it | a simple prediction endpoint | FastAPI or BentoMLmodel serving |
Build on GitHub
Churn model, end to end
A tabular model from raw data to a tracked experiment to a served endpoint, with an honest baseline first.
You are ready to move on when
- Your model beats a simple baseline, and you say by how much.
- Anyone can re-run your best experiment.
- Your endpoint answers in under a second.
Intermediate · 3 months
Walk: You can build it so it keeps working, without help.
Make it production-grade: features, registry, pipelines and safe releases.
Key ideas at this level, the same on every platform:
Learn
| Learn | What it means | On Open source you use |
|---|---|---|
| Feature store | the same features for training and serving | Feastfeature store |
| Model registry | versions, approvals, rollback | MLflow registrymodel registry |
| Pipelines | retrain and test on a schedule | Metaflow or Kubeflowml pipelines Airflow or Dagsterorchestration |
| Safe releases | shadow, canary, rollback | FastAPI or BentoMLmodel serving GitHub Actionsci/cd |
Build on GitHub
Features and serving
Same features for training and live predictions, a model registry, and a canary release with rollback.
You are ready to move on when
- Training and serving features match, and a test proves it.
- You can roll back a model in one step.
- Retraining runs without you.
Advanced · 6 months +
Run: You can design it, prove it and lead others through it.
Run models as a platform: monitoring, drift, cost and many teams.
Key ideas at this level, the same on every platform:
Learn
| Learn | What it means | On Open source you use |
|---|---|---|
| Monitoring and drift | notice quiet decay; alert on it | Evidentlymodel monitoring |
| Retraining policy | release only when it beats the current model | Metaflow or Kubeflowml pipelines |
| Platform for teams | templates, shared features, cost per model | Docker Composeinfrastructure as code Query plans and run timescost |
| From ML to LLM | where classic models still win, and how they combine | Ollama, or a free hosted open modelmodel api Eval Harness Framework or promptfooevaluation |
Build on GitHub
Drift and retrain loop
Watch a live model, detect drift, retrain on a schedule, and only release when it beats the current one.
Robotics data benchmark
A benchmark for the data pipelines that feed robot learning: formats, quality and cost.
You are ready to move on when
- Drift raises an alert before a user notices.
- Each model's monthly cost is known.
- A new team can ship a model with your template.
Learn together
Do it with others
Every project above has a page, a list of open work and a "good first issue" label on GitHub. Pick one at your level, say hello by email, and build it in the open. If a project says needs a lead, that can be you.