Tensorhaus course tracks
[Catalogue — Three Tracks]

Every course
has a clear shape.

Three tracks, each with a defined duration, a stated weekly commitment and a published prerequisite list. No ambiguity about what you are signing up for.

[Methodology]

How the curriculum is structured

[Layer 1]

Mathematical and coding foundations

The Foundations track covers the mathematics that practical ML work actually uses — linear algebra, probability and calculus — alongside Python for data work. Neither is treated as decoration; each concept is introduced where it is needed in the code.

[Layer 2]

Architecture and engineering knowledge

The Deep Learning track builds on layer one to cover network architectures, training at scale, experiment tracking and the engineering that surrounds a trained model. This is where theoretical knowledge becomes production fluency.

[Layer 3]

Applied project from start to deployment

The Capstone turns both layers into a single coherent project: framing a real problem, building a dataset, training and evaluating a model, deploying it and writing a technical report that explains every decision.

Foundations of Machine Learning
[Track 01]

Foundations of Machine Learning

An eight-week evening course covering Python for data work, linear algebra as it is used in practice, gradient descent, regression, classification and honest model evaluation. Written for working adults with some programming background who want the mathematics to feel usable rather than decorative. Roughly eight hours a week, split between live sessions and coursework.

8 weeks
~8 hrs / week
Cohort-based
Evening schedule

What is included

  • Weekly graded exercises with written tutor feedback
  • Code review from a tutor on each submission
  • Private cohort forum — questions answered during the week
  • Written record of course completion on finishing all work

Who this suits

Software developers, data analysts and operations professionals who have some programming background and want to build a solid base in machine learning before moving into more advanced material.

[Track 02]

Applied Deep Learning Track

A sixteen-week track through neural network training, convolutional and sequence models, transformers, fine-tuning, and the engineering around them: data pipelines, experiment tracking, evaluation and cost. Suited to learners who finished a foundations course or work already as software engineers. Twelve hours a week including two live sessions.

16 weeks
~12 hrs / week
2 live sessions / week
GPU credits included

What is included

  • GPU credits for all coursework compute
  • Four assessed projects with tutor code review on each
  • Two live sessions per week with the instructor cohort
  • Written record of course completion on finishing all work

Who this suits

Engineers who have completed a machine learning foundations course or have equivalent practical background and want to move into neural networks, transformers and the tooling around production deep learning systems.

Applied Deep Learning Track
Mentored Capstone and Portfolio Programme
[Track 03]

Mentored Capstone & Portfolio Programme

A twenty-four-week programme in which each learner ships one substantial project end to end — from problem framing and data collection through training, evaluation, deployment and a written technical report. Intended for people who want a body of work to show rather than a list of topics studied.

24 weeks
One-to-one mentoring
Cloud credits included
Portfolio output

What is included

  • Weekly one-to-one mentoring with a practising engineer
  • Fortnightly cohort reviews — present work, get feedback
  • Portfolio preparation and interview practice workshops
  • Cloud credits for training and deployment infrastructure
  • Written record of course completion on finishing all work

Who this suits

Professionals with solid machine learning and engineering knowledge who want to build one substantial, deployed project that demonstrates what they can do in a real work context — rather than a list of completed modules.

[Which track?]

Choosing the right track

The tracks are designed to follow one another but you can enter at the level that fits your current background.

Feature Foundations Deep Learning Capstone
Duration8 weeks16 weeks24 weeks
Weekly hours~8 hrs~12 hrs~12 hrs
Live sessions
Tutor code review
GPU credits
One-to-one mentoring
Portfolio project
Completion record
Price (RM)4802,4004,180

Best for →

Learning the foundations with some programming background

Most popular →

Moving into neural networks and production engineering

Best for →

Building a concrete project to demonstrate existing skills

[Standards]

Shared across all tracks

Data privacy

Learner data is used only for course administration. We comply with the Malaysian Personal Data Protection Act (PDPA) and do not share enrolment information with third parties.

Assessment standards

All assessed exercises are reviewed against a published marking rubric. Feedback is written and specific to the submitted work — not generated from a template.

Curriculum review cycle

Material is reviewed after each cohort and updated where learner questions indicate gaps. The curriculum is not static — it reflects what working AI practitioners actually encounter.

[Enquire]

Not sure which track is right?

Send an enquiry and we will look at your background and let you know which track fits — and what to prepare if the next level is a step away.

Talk to the Team