Subjects / Machine Learning in Python

The applied
Machine Learning subject.

A hands‑on Machine Learning subject for analysts and consultants who already know the basics of Python. Nine courses and projects covering regression, decision trees, k‑means clustering, feature engineering, advanced regression, classification and clustering, with two workplace projects on car pricing and loan default risk.

Data scientist working at a laptop in a modern office, building machine learning models in Python
Curriculum
9 courses & projects
Total Learning
~19 hours
Level
Intermediate to Advanced
Credits
16 CPE/CPD
Try for Free → Explore the curriculum
1M+ learners trained 4.7/5 average G2 rating 80% course completion rate
Trusted by finance and consulting teams at
Why Machine Learning in Python?

Predictive analytics finance
and consulting teams can actually ship.

Python is the language of modern machine learning, and finance and consulting teams are increasingly expected to work with the models, not just the outputs. This subject takes analysts who already know the basics of Python from regression through to advanced classification and clustering, with two workplace projects to lock the skill in.

See the learning outcomes →

Regression to advanced classification

Build a first ML algorithm in week one. By the end, deploy advanced classification with Naïve Bayes and Support Vector Machines, plus advanced clustering for fraud detection and customer segmentation.

Two workplace projects, not just lectures

A pricing strategy project for a Japanese carmaker re‑entering the US market, and a credit‑risk project for a bank identifying likely loan defaulters. Skills practised on real business decisions.

Feature engineering, in detail

A dedicated course on the work that makes models actually useful. Feature scaling, feature selection and dimensionality reduction, taught with applied datasets.

Credentials that stand up

Every milestone is independently accredited by CPD, CPE and NASBA, so what learners earn here counts towards continuing professional development.

Learning Outcomes

What learners will be able to do
by the end of the program.

Every outcome is mapped to a specific lesson and assessed through scenario‑based exercises. Learners walk away with practical ML skills they can apply on the next forecasting, segmentation or risk modelling task.

Build and evaluate regression models

Linear regression to predict numeric values, then advanced techniques (decision tree, KNN, support vector and logistic regression) for the cases linear can't handle.

Regression

Build classification models, basic to advanced

Decision trees as the easy‑to‑visualise starting point, then advanced classification with Naïve Bayes and Support Vector Machines for fraud detection and credit risk.

Classification

Cluster and segment data with confidence

K‑means clustering to start, then advanced clustering (single linkage and soft clustering) for fraud detection and segmenting customers in messy real datasets.

Clustering

Engineer features that improve every model

Feature scaling for uniform inputs, feature selection for predictive power, dimensionality reduction for trimming noise. The work that turns a fragile model into a useful one.

Feature Engineering

Apply ML to a real pricing decision

A workplace project for a Japanese carmaker re‑entering the US market. Use ML to identify which factors should drive pricing so the team avoids the previous over‑pricing miss.

Workplace Project

Apply ML to credit risk modelling

A workplace project for a bank. Use several classification algorithms to separate likely defaulters from customers who'll continue paying back their loans, with the explanation a credit committee actually needs.

Workplace Project
The Curriculum

Nine courses and projects,
in the order learners take them.

Machine Learning in Python is a single applied track. Nine items running from regression through advanced classification and clustering, with two workplace projects on car pricing and loan default risk. Roughly 19 hours end to end.

ML in Python · 1 of 9Advanced

Regression Analysis in Python

Build a first machine learning algorithm in Python. Use linear regression to predict future numeric values from existing data, applied to a real business scenario.

3 hours 2.5 CPE/CPD
ML in Python · 2 of 9Advanced

Decision Trees in Python

Classification algorithms predict outcomes with a few possible variations. Decision Trees are easy to understand and to visualise, the natural starting point for classification.

2.5 hours 2.0 CPE/CPD
ML in Python · 3 of 9Advanced

K‑Means Clustering in Python

Deploy the k‑means clustering algorithm. Segment customers into separate groups so the business can tailor responses to each one.

1.5 hours 1.0 CPE/CPD
ML in Python · 4 of 9Advanced

Feature Engineering

The work that makes models actually useful. Feature scaling for uniform inputs, feature selection for the most predictive power, and dimensionality reduction for trimming noise.

2.5 hours 2.0 CPE/CPD
ML in Python · 5 of 9Advanced

Advanced Regression

Beyond linear. Nonparametric methods including decision tree regression, k‑nearest neighbours regression, and support vector regression, finishing with logistic regression.

1.5 hours 1.5 CPE/CPD
ML in Python · 6 of 9AdvancedProject

Improve a car company's pricing strategy

A workplace project. Help a Japanese carmaker re‑enter the US market. Their previous attempt failed on price, and ML can identify which factors should drive the new strategy.

1.5 hours 1.5 CPE/CPD
ML in Python · 7 of 9Intermediate

Advanced Classification

Detect transaction anomalies and assess loan default risk. Two new algorithms (Naïve Bayes and Support Vector Machines) extend the classification toolkit beyond decision trees.

3 hours 2.5 CPE/CPD
ML in Python · 8 of 9Intermediate

Advanced Clustering

Single Linkage Clustering and Soft Clustering go beyond k‑means. Identify fraud, segment customers more precisely, and handle the messy real datasets simple methods can't.

2 hours 1.5 CPE/CPD
ML in Python · 9 of 9AdvancedProject

Identify risk of default with predictive analytics

A workplace project. Help a bank separate likely loan defaulters from customers who'll keep paying. Use several classification algorithms and arrive at a model the credit committee can defend.

1.5 hours 1.5 CPE/CPD
9
Courses & Projects
~19h
Total Learning
16.0
CPE / CPD Credits
100%
Self‑paced

Machine Learning in Python builds directly on Python Fundamentals. Both sit inside Kubicle's wider library, alongside subjects on AI Fundamentals, Excel, Power BI, SQL, Alteryx and financial modelling.

Try for Free →
The Kubicle Learning Experience

Why learners actually finish
and apply what they learn.

Kubicle is engineered around a single goal: practical skills that get used. Every design choice, from lesson length to assessment style, is made to keep finance and consulting professionals engaged and to translate watch time into measurable outcomes at their desk.

Bite‑sized lessons

Most lessons run 5–10 minutes. Designed to fit between meetings, not block out a Tuesday afternoon. The result: 80% completion vs. a 7% industry benchmark.

Real workplace projects

Learners build a portfolio of project work, not a list of completion certificates. Skills practised on the kind of business decisions ML actually shows up to inform.

Independently accredited

CPD, CPE and NASBA accreditation means credits count toward continuing professional development. Certificates are shareable on LinkedIn the moment they're earned.

Modular and stackable

Every course slots into a wider learning path. Pair Machine Learning in Python with AI Fundamentals, Python Fundamentals or Data Literacy as the team's needs grow.

Manager‑level visibility

Engagement and skill‑gain dashboards from day one. Leaders see exactly where capability is building and where it isn't, by team and by role.

Always current

Modules are refreshed continuously to reflect new tools, regulation and the rise of generative AI. You never pay for material that's gone stale.

Credibility

Why finance and consulting leaders
choose Kubicle, with confidence.

The numbers, the accreditations and the customer outcomes are all on the table, so L&D, Early Careers and Practice Leads can build the business case fast.

Learners trained
1,000,000+
G2 average rating
4.7/5
Course completion
80% · 11x industry
Years in the market
10+ years
Accreditations
CPD · CPE · NASBA
Security
ISO 27001 · GDPR
Independently Accredited

Credentials that count toward CPD.

Every course awards CPE / CPD credits via NASBA‑recognised accreditation, so time spent here counts on a CV and on the team's compliance log.

CPD, CPE, NASBA accreditation logos
Outcomes, Not Hours Watched

Skills learners apply
the same week they learn them.

These aren't aspirational numbers. They are the measured outcomes Kubicle clients report after a typical rollout, with manager‑level dashboards from day one so practice and L&D leaders can see exactly where capability is building.

The Headline

An hour back, every week.

More than 60% of Kubicle learners report saving between 30 minutes and 2 hours every week thanks to sharper data and ML skills. Time redirected to higher-value work, and a measurable productivity dividend at the team level.

60%+
save 30 mins–2 hours every week
Skills That Stick

Used every single day.

45% of Kubicle learners apply newly‑acquired skills daily, the difference between training that's watched and training that's worked.

45%
apply newly‑learned skills daily in their roles
Sharper Communication

Stronger numbers, stronger narratives.

75%+ of learners report measurable gains in both data analysis and the communication skills needed to make insights land with stakeholders.

75%+
gain stronger analysis and communication skills
What Our Partners Say
TotalEnergies
Kubicle was easy to set up and our learners were quickly engaged. It's effectively closed our skill gap, especially in tools like Excel and data processing, and the real-time progress tracking and certifications are why we'll keep using it.

Validated Reviewer
TotalEnergies

Frequently Asked

Answers for the questions we hear most.

Still have a question? Send an enquiry and one of our learning advisors will walk you through it within a business day.

Try for Free →
Who is this subject designed for? +

Finance and consulting analysts and associates who already have a working knowledge of Python and want to extend into machine learning. Roles include risk, FP&A, deals analytics, advisory and corporate finance teams that touch forecasting, segmentation or credit modelling.

How much time does it take? +

Roughly 19 hours of content across 9 courses and projects, worth 16.0 CPE/CPD credits. Most learners complete the full subject in four to six weeks of casual study.

Do I need a Python background first? +

Yes. This subject assumes a working knowledge of Python (variables, types, functions, loops, plus Pandas and NumPy). Learners new to Python should start with the Python Fundamentals subject before this one.

Does this cover regression, classification and clustering? +

Yes, all three. Regression includes linear, decision tree, k‑nearest neighbours, support vector and logistic regression. Classification includes decision trees, Naïve Bayes and Support Vector Machines. Clustering covers k‑means, single linkage and soft clustering. Plus a dedicated course on feature engineering.

Are the credentials accredited? +

Yes. Kubicle credentials are independently accredited by CPD, CPE and NASBA. Every course and project awards CPE/CPD credits that learners can apply toward continuing professional development. Certificates are shareable on LinkedIn the moment they're issued.

Will it integrate with our LMS and SSO? +

Yes. Kubicle offers off‑the‑shelf integrations with leading LMS providers and the engineering team will build a custom integration for any other system you run. SSO is supported across the board, and a Reporting API exposes engagement and completion data to your dashboards.

How is Kubicle secured? +

Kubicle is ISO 27001 certified, GDPR‑compliant, and undergoes regular third‑party penetration testing. Granular roles, SSO and audit logging are built in.

Can we get a tailored quote and pilot? +

Of course. Send an enquiry and a learning advisor will scope the rollout, recommend the right starting point for your team, and put together a pilot proposal with a tailored quote.

Make an Enquiry

Give your team the ML skills they need for the next decade of work.

Send a short enquiry and a learning advisor will come back within one business day with a tailored recommendation: a curriculum mapped to your team's roles, a sample pathway through Machine Learning in Python (and any related subjects), and a quote you can take to budget. Built for teams of 5 to 100, deployable in days.

Tailored curriculum recommendation for your team Sample learner pathway by role Volume-based quote, no obligation Reply within one business day
Try for Free →
Visit
kubicle.com
Email
sales@kubicle.com
Offices
Dublin, Ireland
Talk to Sales →