Three Tracks, One Direction: Practical AI Skill
Deep learning fundamentals, computer vision, and applied machine learning — each track scoped and structured for people who want to build skills they can actually use.
Back to HomeHow the Tracks Are Built
Each track starts from a defined scope. Before writing a module, the curriculum team maps out what a student should be able to do at the end of it — what they'll build, what tools they'll use, and what results they should expect to see.
Content is written with that output in mind, not backward from a list of topics. Exercises are built around real conditions — datasets that aren't pre-cleaned, model parameters that aren't pre-tuned, and pipelines that don't always work the first time.
Quality review happens before release and on a rolling basis afterward. When students flag something that isn't clear, it goes into the review queue. The goal is a track that holds up to repeated use, not one that looks good on first glance.
Output-first design
Each module is defined by what students build, not by topics covered.
Real conditions
Exercises use datasets and setups that reflect actual field work.
Rolling review
Content is updated when the tooling or student feedback warrants it.
Clear sequencing
Tracks state their starting prerequisites and recommended order explicitly.
Deep Learning Specialization
A focused track on neural network design, training techniques, and model tuning with applied exercises. For students who have covered the fundamentals and want depth. This track goes beyond running a pre-written training loop — you'll write your own, examine what happens when it fails, and rebuild it in a way you understand.
How This Track Works
Module scope review — read what you'll build and what tools you'll need
Work through written material and code walkthroughs at your pace
Complete the module exercise — includes at least one failure-case task
Post questions in the community or support channel, then move to the next module
Computer Vision Intensive
A project sprint covering image data, model training, and evaluation for visual tasks. Practical and outcome-focused across a defined timeframe. The structure is deliberately sprint-format — modules move faster, the scope per week is higher, and the expectation is that you come in with existing ML knowledge and want to add computer vision work to it.
Sprint Structure
Week 1–2: Image data handling and preprocessing foundations
Week 3–4: Model architecture selection and training setup
Week 5–6: Evaluation, iteration, and project documentation
Applied Machine Learning Track
Project-led learning where students build, evaluate, and document real models on practical datasets. Designed for learners ready to move from theory into hands-on work. This is the most end-to-end of the three tracks — you'll work from raw data through to a documented, reproducible model output, with all the messy middle parts included.
Project Workflow
Data inspection and problem framing — what kind of ML task is this?
Feature pipeline design and baseline model implementation
Systematic model iteration and evaluation against held-out data
Model documentation and project write-up with reproducibility notes
Which Track Fits You?
The tracks are designed to complement each other. Here's how they compare so you can pick the right starting point.
| Feature | Deep Learning ฿4,200 |
Computer Vision ฿5,950 |
Applied ML ฿11,900 |
|---|---|---|---|
| End-to-end project workflow | Partial | Partial | |
| Neural architecture depth | CV-focused | Overview | |
| Image data and pipelines | — | — | |
| Feature engineering focus | — | — | |
| Sprint-format delivery | — | — | |
| Model documentation skills | — | — | |
| Failure-case exercises | |||
| Best for… | Understanding neural networks inside out | Adding computer vision to existing ML skills | Moving from theory to full project capability |
Not sure which to start with? Send us a message — we'll help you figure it out.
What Every Track Shares
Data Privacy
Student data is used only for enrolment and communication. No third-party sharing without consent.
Technical Accuracy Review
All module content is reviewed by practitioners before publication and updated on a rolling basis.
1–2 Day Support Response
Track-related questions receive human responses within 1–2 business days on all support channels.
Open-Source Tooling Only
All tracks use freely available Python libraries. No proprietary platform subscriptions required.
Module Scope Documents
Every module begins with a scope document so students know what they're committing to before they start.
Moderated Community
A shared student community for questions and discussion, kept focused and technically useful.
Straightforward Track Pricing
One payment per track, no subscription, no hidden extras. Payment details are provided after you submit an enquiry.
Deep Learning
Specialization
per track · one-time
- Full track access
- Community access
- Email support
- Content updates
Computer Vision
Intensive Sprint
per track · one-time
- Full track access
- Community access
- Email support
- Content updates
Applied ML
Full Track
per track · one-time
- Full track access
- Community access
- Priority email support
- Content updates
Not Sure Where to Start?
Send us a message describing your background and what you want to be able to do. We'll suggest the track that makes sense for where you are right now.
Get a Track Recommendation