What makes learning here
different from elsewhere.
We made a set of deliberate choices about how courses run. This page explains what those choices are and why they shape a better learning experience.
Back to HomeSix things worth knowing.
Instructors with field experience
Both lead instructors worked in data and software roles before joining Tunas AI. The material they teach reflects what the work actually looks like.
Cohort size kept small
We limit intake so that clinic sessions can be genuinely useful. When questions come up, there is time to work through them — not just acknowledge them.
Written feedback on every project
Projects are reviewed by the instructor personally. Feedback addresses specific decisions in the learner's code, not templated comments.
A connected three-track path
The three courses form a coherent sequence. Each track is self-contained, but each one also prepares learners well for the next if they choose to continue.
Practical, not theoretical
Each week closes with something built — a working script, a trained model, a deployed endpoint. Learning is measured in things made, not pages read.
Designed for working adults
The weekly workload across all tracks is shaped for people with full-time jobs. The pace is steady, not frantic. Session recordings mean missing a week is recoverable.
Each benefit, explained.
Professional Expertise
The instructors at Tunas AI are not career academics who moved into teaching after completing research. Ahmad Zulkifli spent six years in a data engineering role in the fintech industry. Siti Rohani completed her research on neural architectures and then worked on production ML systems before joining the school. That background shapes how they explain concepts — with reference to what matters in practice, not just what is cleanest in theory.
- Industry-background instructors on both Python/ML and deep learning tracks
- Curriculum reviewed each intake to reflect the current ecosystem
- Projects use tools and patterns relevant to the local market
Technology and Tooling
Tunas AI courses are built around tools learners will encounter after the course ends: Python with standard data science libraries, scikit-learn for machine learning practice, and PyTorch for the deep learning track. We do not build proprietary sandboxes or require learners to work in limited environments. The setup you use in the course is the setup you carry forward.
- Standard open-source toolchain throughout — no locked environments
- Track 03 includes an introduction to deployment alongside training
- Sessions recorded on a private platform accessible throughout the course
Learner Support
Questions that arrive outside the weekly clinic are handled through the peer channel and email. The programme coordinator aims to respond to administrative enquiries within one working day. For technical questions, the instructor reviews and responds before the next clinic session.
- Weekly live clinic for direct questions to the instructor
- Peer channel for day-to-day discussion between cohort members
- Email support for enrolment and administrative matters
Value and Pricing
Pricing at Tunas AI reflects the size of each track and the level of instructor involvement. Track 01 is RM 960 for six weeks, Track 02 is RM 1,410 for ten weeks, and Track 03 is RM 1,870 for fourteen weeks. Each fee covers the course content, clinic sessions, project feedback, recordings, and peer channel access. There are no additional charges for standard course materials.
- Flat fee per track — no hidden add-ons
- Payment options discussed at enquiry stage
- Refund and deferral conditions outlined before payment
Outcomes
Learners who complete any Tunas AI track leave with at least one project they built themselves — not a walkthrough they followed. By the end of Track 03, the capstone project represents a full cycle of development: problem definition, data work, model training, evaluation, and deployment. That is the kind of portfolio piece that stands up in a technical conversation with an employer or collaborator.
- Each track closes with at least one independently built project
- Track 02 includes two projects for broader portfolio coverage
- Track 03 capstone covers the full development and deployment cycle
Tunas AI vs typical online AI courses.
| Feature | Typical Online Courses | Tunas AI |
|---|---|---|
| Cohort size | Hundreds of learners | Small, capped groups |
| Project feedback | Automated or peer-only | Instructor-written feedback |
| Live instructor access | Rarely or forum-only | Weekly clinic every track |
| Structured progression | Choose your own path | Three connected tracks |
| Tooling used | Proprietary sandbox | Standard open-source stack |
| Post-course community | None or inactive | Alumni channel (Track 03) |
Distinctive features of the Tunas AI approach.
The loft learning model
Our "inference loft" philosophy means courses are structured but not cramped. Each week has a clear task and a clear deliverable, with space between sessions to process and practise before the next step arrives.
Malaysia-contextualised examples
Project datasets and use cases reflect industries active in Malaysia — logistics, banking, retail, and public services — so learners are working with problems they may actually encounter.
Curriculum reviewed every intake
The Python and ML ecosystem changes quickly. We review the course content before each new cohort and update examples or tools where the field has moved on since the last intake.
Side-desk design philosophy
Our course materials are structured so that code and notes sit alongside each other — not in separate tabs or separate documents. This makes review and reference more natural during and after the course.
Milestones and standing.
Ready to see which track fits?
Send an enquiry and we will explain the options, share current intake dates, and answer anything you need before deciding.
Send an Enquiry