Machine Learning A–Z
Signature Course No. 212 — A Signature AI Socratic Coaching Course.
A New Kind of Learning
Socratic coaching, delivered by an agentic AI coach.
Not a course you watch. An AI coach you work with.
The Lighthouse Coach begins by discovering how you think about the subject, teaches what is missing, and gives you realistic practice before asking you to prove that you can use it. You may test out of concepts you already understand, while capstones turn the learning into a practical plan, decision, conversation, or other real-world result.
The curriculum is deliberately authored; the AI delivers it through coaching. It is not an AI randomly generating lessons.
A coach that teaches by asking — not by lecturing
The Lighthouse Coach is an AI Socratic coach built on Cognosco Learning's Machine Learning A–Z course, in the AI at Work category. It does not play videos at you. It holds a conversation: it asks you a question, listens to your answer, names the misconception it hears, affirms what you got right, and asks the next question — the way a great coach works in the room.
Machine learning is a way of building answers out of examples. Instead of writing rules by hand, you show a procedure many past cases and let it adjust itself until its answers on those cases are good — then comes the only question that matters: will it be good on the next case, the one it has never seen? This course teaches the whole arc of that idea, end to end.
You learn to frame before you fit. Prediction versus description, supervised versus unsupervised, regression for numbers and classification for categories — and the discipline that keeps them honest: fitting on one portion of data and judging on another the model was never allowed to see, with every score computed from printed counts you can recompute yourself.
Then the survey widens. Unsupervised methods that find groups and structure without labels, clustering, dimensionality reduction, decision trees, ensembles that combine many models into a stronger one, and neural networks at survey depth — what their layers do, what depth buys, and what it costs — closing with limits: what these methods cannot know, and what a human decision-maker still owes the people a model's answers will touch.
It works in two movements. First, Module 1 — The Honest Fit: Framing, Supervised Learning, and How Results Are Judged — work through Concepts 1–9, finishing with The Audit Before Anyone Fits Anything; then, Module 2 — The Wider Survey: Structure, Combinations, Networks, and Limits — work through Concepts 10–18, finishing with The Defended Choice.
Both capstones produce real artifacts — The Audit Before Anyone Fits Anything, a complete framing-and-evaluation audit of a proposed model, the document a decision-maker should demand before any fitting begins; and The Defended Choice, a defended method-selection and limits statement for a harder, mixed problem, with every score recomputed and every limit confessed.
The numbers discipline is part of the course. Every figure recomputes from its printed counts: where a score is stated, the counts it divides are printed beside it, and no benchmark figure, leaderboard score, or famous-dataset result is quoted anywhere — a survey course that borrows authority from numbers it does not show has taught its first bad habit.
The course is self-paced and always on — roughly 30 minutes per concept, about 9 hours in full, taken in as many sittings as you like. You can type your answers, or talk out loud: the Voice Coach speaks its questions and listens to your spoken replies. Learners who already know the material can test out concept by concept.
Machine Learning A–Z is part of The Lighthouse Signature Series — the online counterpart to the one-day, in-person learning Cognosco Learning has delivered since 2002.
Course Facts
One full concept, coached end to end — free: verify your email and we’ll send you a link to start.
Already enrolled? Enter the course →
Diagnose → Learn & Practice → Prove → Capstone
Every learner moves through the same discipline a Cognosco instructor would insist on in the room.
Diagnose
The Coach starts where the course starts: with what you actually bring. A model fitted before its problem is framed is an answer in search of a question — and a score earned on data the model has already seen is no score at all. Every concept opens with three questions that locate where you actually stand, and your answers shape how deeply the Coach works the concept with you. Everyone begins with Module 1 — no skipping the foundations you only think you have.
Learn & Practice
Work each concept in conversation with the Coach — spoken or written — then practice it against realistic situations until it holds up.
Prove
Every concept ends in proof: answer the check cold and demonstrate the skill. Know it already? Test out and take credit immediately.
Capstone
Finish with two capstones — The Audit Before Anyone Fits Anything for Module 1 and The Defended Choice for Module 2 — that pull the whole course together, and earn the certificate.
What You'll Learn
Framing comes first. Prediction versus description — an estimate graded by hidden truth versus structure judged by usefulness — decided before any method is chosen, because a strong model of the wrong problem is this field's most common failure.
Module 1 teaches the honest fit. Supervised learning from labeled examples, regression and classification run by hand, the split that keeps fitting and judging apart, both scoring languages computed from printed counts, and features audited for legitimacy and leakage.
Module 2 widens the survey. Unsupervised learning, clustering, dimensionality reduction, trees, ensembles, and neural networks at survey depth — what each family is for, where each breaks, and what depth buys and charges.
And it draws the border. Leakage, drift, and skew named and audited; models chosen against honest baselines under one fair protocol; and every project closed with a limits statement — territory fenced, construct confessed, and a named human who decides.
What you will learn
- Frame a machine-learning problem — prediction versus description, supervised versus unsupervised — before choosing any method, with unit, grading fact, and decision-owner named
- Split data honestly, fit on one portion, and judge only on data the model never saw — and score numbers with error measures and categories with confusion counts, precision, recall, and accuracy, each computed from printed counts
- Recognize overfitting, design features a model may legitimately see, and catch leakage — information from the future smuggled into the past
- Survey clustering, dimensionality reduction, trees, ensembles, and neural networks — what each family is for and where each breaks — then choose against an honest baseline and state any model's limits in plain words
Everything in the course
- All 18 concepts of Machine Learning A–Z, in 2 modules
- A starting-point check first — the Coach finds where you actually stand before teaching
- Learn, Practice, Prove for every concept — no passive video watching
- A test-out on every concept: prove you know it and move on
- The Voice Coach: talk with the Coach out loud — it speaks, listens, and replies
- A personal Learner Record of your work across the course
- Two capstone exercises: The Audit Before Anyone Fits Anything and The Defended Choice
- A printable session summary whenever you pause
- The Lighthouse Coach completion certificate (Signature Series)
- Your progress is saved as you go — stop anytime, resume where you left off
Built for anyone who must judge a model's claims
Managers and decision-makers — who are pitched machine-learning projects, shown impressive scores, and asked to fund or approve them, and who need to know which questions expose a sound project and which expose a demonstration.
Analysts, builders, and the simply curious, with standards — analysts and data-adjacent professionals who want the full map of the field with honest measures before they specialize; developers, operators, and founders who will choose methods, prepare data, and defend results; and anyone who wants to understand what these systems actually do, beneath the headlines, well enough to argue about them accurately.
Anyone tired of confident figures they cannot check — every dataset in the course small enough to print is invented for teaching and labeled as such, every derived figure recomputes from printed counts, and the teaching is tool-neutral, assuming only comfort with ordinary arithmetic and no programming experience.
Prefer to learn live and in person? The same material anchors Cognosco's one-day programs — browse the in-person catalog.
Start with The Audit Before Anyone Fits Anything
$195 — introductory price. One payment. No subscription, no password to manage.
Here is exactly what happens when you enroll:
- Pay securely with Stripe — you are taken straight into the course
- Create your learner profile and begin the Module 1 starting-point check
- Save the access key you are given — it restores your enrollment on any device
- Learn at your pace; your progress is stored in your browser on your device
For the full voice experience, use Chrome and allow the microphone when asked. Typing your answers works everywhere, all the time. Questions before you enroll? Email hello@cognoscolearning.com.
More Signature Courses are coming
Machine Learning A–Z is Signature Course No. 212. Leave your email and we'll send you news of future Signature Courses and Lighthouse updates as they are announced — nothing else.