MLOps engineer
Learn to deploy, monitor and retrain machine learning models and AI services in production, working with AI coding tools from week 1, in 20 weeks.
Coming soon · planned, not written yet
- 20
- weeks planned
- ~40 h
- a week
- 5
- milestone projects
- 1
- project, grown every week
These are planned figures. The weeks are written after we check the tools against current job postings and official documentation, so the final route may differ.
Built for the AI era
How AI fits this roadmap
AI coding assistants now draft much of the pipeline code, Dockerfiles and config that MLOps engineers write, so the job leans toward reviewing, testing and owning what ships. This roadmap has you use an assistant every week and check its output with tests, linters and plan reviews. Teams also now run language model services next to classic models, so you serve, trace, evaluate and monitor both.
The main project
ModelLine
A prediction service plus a small language model feature that you take from a notebook to a tracked, tested, deployed, monitored and retrainable production setup.
What you will be able to do
- 01
Turn a notebook model into a tested Python package, using an AI assistant to draft code and tests while you read and fix every line.
- 02
Track experiments and register versioned models with MLflow, so any result can be reproduced.
- 03
Build a CI/CD pipeline that tests, packages and deploys a model service as a Docker container to Kubernetes.
- 04
Serve a language model behind an API, trace its calls and score its answers against a fixed test set before each release.
- 05
Monitor data drift, latency, errors and cost with Evidently, Prometheus and Grafana, and trigger a retrain when a check fails.
01The route
Five months, one month at a time.
Each month ends in a milestone project, so you always have something to show for the work.
Month 1 · Weeks 1–4
Python, Git, containers and AI-assisted work
Packaging Python, tests, Git workflows and Docker. You set up an AI coding assistant, learn to prompt it, read its diffs and test what it writes.
MilestoneA trained model from a notebook, rebuilt as a tested package in a container, with a log of AI suggestions you accepted and rejected
Month 2 · Weeks 5–8
Experiments, data and model registry
Experiment tracking and a model registry with MLflow, versioned data and repeatable training runs. You use AI to draft training code and then check it for leaks and errors.
MilestoneA training pipeline whose every run is tracked and whose best model is registered
Month 3 · Weeks 9–12
CI/CD and serving on Kubernetes
GitHub Actions for tests and builds, model serving with BentoML, and deployment to Kubernetes with rollbacks. You use AI to draft pipelines and manifests, then fix what is wrong.
MilestoneEvery merge tests, builds and deploys the model service, and a bad release can be rolled back
Month 4 · Weeks 13–16
Serving and evaluating language models
Serve an open language model with vLLM, call hosted models through an API, trace calls with MLflow and score answers against a test set. You learn where model output cannot be trusted.
MilestoneA language model feature in ModelLine with traces and an evaluation check that blocks a bad release
Month 5 · Weeks 17–20
Monitoring, retraining and capstone
Drift and quality checks with Evidently, metrics and alerts with Prometheus and Grafana, automated retraining, cost control, a portfolio write-up and interview preparation.
MilestoneA live dashboard, an alert that triggers a retrain, one incident review and a portfolio write-up
03What is inside
Everything a week comes with.
Every week is built the same way, so once you have done one you know how all of them work.
- 01
One project, ModelLine
You grow one model service from notebook to production, with a language model feature inside it, so every tool lands in something real.
- 02
A day-by-day plan
Each week has a Learn part and a Do part, with a checklist so you know when you are done. Most weeks include a task to do with an AI assistant and a step to check its work.
- 03
A project in every milestone
Five milestones, each ending in something you can show and link to.
- 04
Reference topics
Short topic pages with check questions you can come back to.
- 05
A job track
Interview practice, portfolio and resume help, and an application plan.
- 06
Progress in your Study plan
Add weeks to your Study plan, give them dates, and finished days are counted for you.
04Is it for you
Who this is, and is not, for.
Better to know before you start than six weeks in.
It is for you if
- You can write Python and have trained a simple model, or you are a DevOps or backend engineer moving toward ML.
- You want to be the person who makes models and AI features run reliably, and you want to be strong with AI tools.
- You can give about 40 hours a week for five months.
It is not for you if
- You have never written code or used a terminal. Do a beginner course first.
- You want a job guarantee. Nobody can promise that, and this roadmap does not.
- You want to research new model architectures or train large models from scratch. This roadmap is about running them.
- You want to paste AI output into production without reading it. This roadmap trains you to check it.
?Questions
Before you start
- Is this roadmap ready?
- Not yet. This is the plan. The weeks are being written and checked against current job postings and official documentation. Leave your email and we will write to you when it opens.
- What will it cost?
- The price is not set. Roadmaps are bought once, with no subscription. We will announce it when this one opens.
- Will the plan change?
- It may. Before the weeks are written we check the tools against current job postings, so the lengths and months here are planned, not final.
- Do I need a GPU or a cloud account?
- Not for most weeks. Classic models run on a laptop. The language model weeks will show a small model that runs locally or a hosted API with a low bill, and will say what each option needs once the weeks are written.
- Will it get me a job?
- It will give you skills and proof of work. It cannot promise a job on a fixed date, because hiring depends on the market, your location and your past work.
Not open yet
Leave your email and we will write once, when it opens.