Three courses, each with
a published syllabus
From Python fundamentals for numerical computing to a supervised employer engagement, each programme is described in full before you decide whether to enrol.
// methodology
How the courses are structured
Sequential by design
Python for ML Practice feeds into Deep Learning Systems. The entry course covers the array and data-handling foundations that the advanced course assumes. Participants can enter Deep Learning Systems without the entry course if the readiness check confirms the background.
Assessment defined in advance
Each course documents what will be assessed, by what criteria, and when, before the first session. Graded notebooks use a rubric shared at the module start. The lab report format is the same across all four reports in the deep learning course.
Cohort as a category, not an upgrade
The employer cohort is not a more expensive version of the individual courses. It is a different product — the deep learning syllabus delivered alongside supervised work on an internal problem. It requires a scoping engagement and a defined internal codebase.
Python for Machine Learning Practice
An eight-week entry course for people who write some code but have not worked with numerical libraries. The focus is on understanding what the tools are doing, not only on using them. Coverage includes NumPy array semantics, pandas dataframes and their common failure points, vectorisation and why it matters, plotting for diagnosis rather than decoration, notebook discipline, environment and dependency management, and how to test numerical code.
Topics covered
- NumPy array semantics and broadcasting
- pandas dataframes — indexing, merging, common pitfalls
- Vectorisation and loop avoidance patterns
- Plotting for diagnostic use, not decoration
- Environment management and reproducible notebooks
- Testing numerical code
Take the readiness quiz
Published quiz, taken online. Result shared with you; school gives plain feedback on fit.
Confirm enrolment and pay fees
Invoiced directly; course materials and schedule shared after payment confirmed.
Weekly sessions + graded notebooks
Three contact hours and five exercise hours per week across eight weeks.
Data-cleaning project and record of completion
Short project in final weeks. Written completion record issued on passing assessment.
Confirm prerequisite background
Via completion of the entry course or a separate readiness confirmation for experienced practitioners.
14 weeks of sessions and lab work
Six contact hours and ten to twelve lab hours weekly. Compute sandbox provided from week one.
Four laboratory reports
Each graded against a rubric shared at the module start. Reports document design decisions, not just results.
Published-result reproduction task
Final assessment: reproduce a paper's training result, document deviations and their causes.
Deep Learning Systems
A fourteen-week course on building and training neural networks with intent. The course works through the reasoning behind each major design choice rather than treating architectures as black-box tools. Topics include backpropagation from first principles, initialisation and normalisation, convolutional and attention-based architectures, data pipelines, mixed precision, distributed training across a small cluster, profiling, reproducibility practice, evaluation design, and documented model limitations drawn from published research.
Topics covered
- Backpropagation from first principles
- Initialisation, normalisation, and their effects
- Convolutional and attention-based architectures
- Mixed precision, distributed training, profiling
- Reproducibility practice and evaluation design
- Documented model limitations (sources cited by name)
Employer Cohort with Project Supervision
A twenty-week engagement combining the school's deep learning syllabus with supervised work on an internal problem the employer brings. The engagement begins with a scoping workshop with engineering leadership to define the problem scope and expected outputs. Weekly cohort sessions for up to twelve engineers run alongside fortnightly code review against the employer's own repository. A written architecture review is produced at the midpoint. The engagement closes with supervised deployment to the employer's staging environment, documentation handover, and two clinics after completion for questions arising from early production use.
The school supervises engineering practice and does not take responsibility for the employer's production systems. Priced per cohort; the figure above reflects a standard twelve-seat engagement. Non-standard scope is priced after the scoping workshop.
Scoping workshop
Define problem scope, codebase access, expected outputs, and team composition with engineering leadership.
Weekly sessions + fortnightly code review
Cohort sessions deliver the DL syllabus. Code review is against the employer's actual repository, not sample data.
Midpoint architecture review
Written review of design decisions made to that point. Shared with engineering leadership and team.
Staging deployment + documentation handover
Supervised deployment to staging. Documentation covering architecture, decisions, and limitations.
Two post-completion clinics
Structured sessions after the engagement closes for questions arising from early use. Scope defined at handover.
// which_course
Choosing the right course
Each course targets a different experience level and situation. The comparison below helps identify the right starting point.
| Criteria | Python for ML Practice | Deep Learning Systems | Employer Cohort |
|---|---|---|---|
| Duration | 8 weeks | 14 weeks | 20 weeks |
| Weekly commitment | ~8 hrs | ~16–18 hrs | Team-defined |
| Entry point | Readiness quiz | Entry course or equivalent | Scoping workshop |
| Compute provided | Employer env | ||
| Suitable for | Individual, numerical coding background | Individual, post-entry or equivalent | Engineering team, real problem |
| Price (RM) | 980 | 3,200 | 4,700 / cohort |
// protocols
Standards shared across all courses
Data under PDPA 2010
Enrolment and assessment data processed under Malaysia's Personal Data Protection Act 2010. Retention schedules in Privacy Policy.
Rubric shared at module start
Assessment criteria for each graded piece are shared at the beginning of the relevant module. No criteria are withheld until submission.
Citations by paper name
All references to published research are cited by author and title. Participants can read the source material directly.
Response within one business day
Enquiries sent to [email protected] or via the contact form are acknowledged within one business day (Monday–Friday, KL time).
Withdrawal terms before course starts
Refund terms are documented in the Terms & Conditions before enrolment. Partial refunds are available before a specified point in each programme.
English instruction, BM on request
All courses are delivered in English. For employer cohorts, Bahasa Malaysia delivery can be discussed during the scoping workshop.
// pricing
Course fees
All prices are in Malaysian Ringgit (RM). Listed prices are the full fee; no add-on charges for compute or materials.
Entry level
Python for ML Practice
RM 980
per participant / 8 weeks
- Published syllabus
- Readiness quiz included
- Graded notebook assessment
- Written completion record
Advanced
Deep Learning Systems
RM 3,200
per participant / 14 weeks
- Published syllabus
- GPU compute sandbox included
- 4 graded lab reports
- Published result reproduction
Employer engagement
Employer Cohort
RM 4,700
per 12-seat cohort / 20 weeks
- Scoping workshop included
- Fortnightly code review
- Architecture review + documentation
- 2 post-completion clinics
Questions not answered in the syllabus?
Send a message with your current background and which course you are looking at. We'll reply within one business day.
Get in Touch