Also posted as ML Engineer · Applied ML Engineer · Applied Machine Learning Engineer · MLE
Fine-tunes, evaluates and ships models a product depends on, and proves each change with numbers.
This is the LLM-era version of the job: prompts as a baseline, preference data, fine-tuning and RL, and the cost of running what you trained. Fine-tuning builds on evaluation, which builds on having an agent to evaluate, so the path is longer than a job spec would suggest. It starts with the classical machine learning and statistics every later course assumes, and ends with serving the model you trained.
11 courses88 lessons~263 hours
What the job needs
Solid: course readyDashed: coming soon
Every requirement on this path has a course. Nothing to wait for.
11
Courses
0
Coming soon
~263
Hours
Every course on this path assumes you can write Python or TypeScript and work in a terminal. These foundations aren't written yet, so they sit outside the course count. Tick what you already know.
4 to learn · ~36h · 0 known
Syntax, packages with uv, reading API responses and the parts of the standard library you'll actually use.
Planned~12hUsed by 5 courses
Types, async and await, npm, and running a Node 24 project from the terminal.
Planned~12hUsed by 3 courses
Shell basics, branches, pull requests and the commands Claude Code runs on your behalf.
Planned~6hUsed by 8 courses
Containers, Compose, env files and the sandboxes the later courses run their exercises in.
Planned~6hUsed by 1 course
Learned these elsewhere? The setup guides get Claude Code, Python with uv and Node with TypeScript running in about an hour, so you can start course one today.
Setup guidesEach requirement is covered by a course, placed after the ones it builds on. Overviews are open to everyone; sign in to start the lessons.
The classical machine learning, statistics and probability the LLM courses assume, learned by building a ticket-triage model you can defend with numbers.
3 modules · 10 lessons ~26h 10 quizzes Capstone
Modules
Capstone: Ticket triage you can defend
How large language models are built, trained and run, so you can reason about their behaviour, cost and failure modes in production.
3 modules · 9 lessons ~18h 9 quizzes Capstone
Modules
Capstone: Choose a model for document extraction
What an AI agent really is, how it uses tools and context, and how to make one dependable enough for high-stakes work.
2 modules · 6 lessons ~18h 6 quizzes Capstone
Modules
Capstone: A document review agent in Claude Code
Build real agents on production harnesses: Claude Code, the Claude Agent SDK and Vercel's eve, with skills, subagents and human approval built in.
2 modules · 6 lessons ~20h 6 quizzes Capstone
Modules
Capstone: A drafting agent with human approval in eve
Turn "it seems better" into numbers you can defend: error analysis, code graders, calibrated LLM judges and honest statistics.
2 modules · 7 lessons ~20h 7 quizzes Capstone
Modules
Capstone: A full eval suite for your drafting agent
Ask a question, design the experiment, annotate the data, doubt the result, write it up and get it reviewed: the research discipline for AI agent work.
3 modules · 10 lessons ~26h 10 quizzes Capstone coming soon
Modules
Capstone: A small study on the triage agent (coming soon)
Treat prompts, examples, tool descriptions and skills as versioned, measured artefacts, and improve them by hand, by optimiser and in production without breaking what already works.
2 modules · 7 lessons ~17h 7 quizzes Capstone
Modules
Capstone: Optimise and ship the questions drafter
Collect, measure and trust the judgements of reviewers and users: study design, rater agreement, preference rankings, real-use signals and quality estimates from a few labels.
3 modules · 7 lessons ~28h 7 quizzes Capstone
Modules
Capstone: Build a feedback system for a support assistant
Train a small open model for one agent step with supervised fine-tuning, preference optimisation and reinforcement learning, and prove with your eval suite whether it beats an optimised prompt.
3 modules · 8 lessons ~38h 8 quizzes Capstone
Modules
Capstone: Train a triage model, then decide whether to ship it
Know what each case costs in tokens, then cut that cost with caching, trimming, batching, routing, distillation and self-hosting, without losing quality.
2 modules · 8 lessons ~20h 8 quizzes Capstone
Modules
Capstone: Cost-engineer an agent pipeline
Put the model you fine-tuned behind a production endpoint: inference engines, GPU sizing, quantisation, latency tuning, structured outputs, rollout and operations.
3 modules · 10 lessons ~32h 10 quizzes Capstone
Modules
Capstone: Ship the triage endpoint
This path is what the role asks for, mapped to the courses we have. A requirement we haven't written the course for yet stays on the path as coming soon, so you can see the whole job, not just the part we cover.
