Tensorhaus
Notebook open on a desk with handwritten notes
Why Tensorhaus

Structure and feedback — not just content.

There is a lot of AI content online. What is harder to find is a place that gives you a clear sequence, real exercises, and someone who reads your work and tells you specifically what to improve.

Back to Home
Overview

Six things that set our programmes apart.

Feedback on your actual notebooks

Instructors read your submitted work and return specific written comments — not generic automated grading.

A deliberate sequence

Three programmes that connect: data foundations, then machine learning, then deep learning. Each builds on the last.

Live mentor sessions

The ML Pathway and Capstone include video office hours. You can show your screen, share your code, and get help with specific problems.

Designed for working schedules

All lessons are recorded. Study at 23:00 or on Sunday afternoon — the material waits for you.

Portfolio projects you own

Your project work is yours to keep and publish. Instructors help you produce something you can show to others, not just pass an assessment.

Small cohort sizes

Intakes are capped so instructors are reviewing a manageable number of notebooks at a time — not hundreds simultaneously.

Expertise

Instructors who work in the field.

All Tensorhaus instructors continue to work on applied AI projects outside their teaching roles. That means the feedback you receive reflects current practice, not just academic material. When an instructor points out that your approach to feature scaling will cause problems in a production pipeline, they know this from having built and maintained production pipelines.

  • Instructors carry active projects in data engineering, ML, and deep learning research
  • Curriculum is updated each intake to reflect changes in tooling
  • Feedback addresses both correctness and practical judgement
Technology

Notebooks that reflect current practice.

Tensorhaus programmes use Python, pandas, scikit-learn, and PyTorch — the same stack used in applied ML work in 2024 and 2025. Exercises are built as Jupyter notebooks, not proprietary learning platforms. You run the code on your own machine, which means you understand what is actually happening and build the habit of working in a real environment from day one.

  • Standard Python toolchain — no proprietary software required
  • Notebooks updated to current library versions each term
  • Capstone includes a small-scale deployment component
Support

More than access to content.

Enrolling in a Tensorhaus programme means you have someone to ask when you are stuck. Written feedback on submitted notebooks, video office hours in the ML Pathway and Capstone, and responsive email support from the programme coordinator. The aim is to make sure questions that would otherwise stop your progress get answered.

  • Written feedback on two submitted notebooks in Data Foundations
  • Weekly video office hours in the ML Pathway
  • Three mentor code reviews and a panel presentation in the Capstone
Value

Clear pricing, no hidden extras.

Each programme is priced as a single amount in Thai Baht. That price includes all lessons, all notebooks, and all the feedback and mentor time described in the programme outline. There are no add-on modules, no certification fees charged separately, and no subscription that continues after the programme ends.

  • Data Foundations: ฿7,800 — all-in
  • ML Pathway: ฿22,000 — all-in including mentor hours
  • Deep Learning Capstone: ฿34,500 — all-in including panel review
Outcomes

Something you can show at the end.

The goal of each programme is that you leave with something concrete. After Data Foundations, you have completed exploratory data analysis projects you can refer to. After the ML Pathway, you have three small portfolio projects. After the Capstone, you have a substantial deep learning project, documented and deployable, that you presented to a panel and can walk through in detail with anyone who asks.

  • All project work stored in your own repository
  • Capstone covers full pipeline: data → model → deployment
  • Panel presentation develops the skill of explaining your own work
Comparison

How this differs from the alternatives.

Feature Typical video platforms Tensorhaus
Written feedback on your work
Live mentor video sessions
Structured progression across programmes
Small cohort with instructor attention
Portfolio project you own and can publish sometimes
Self-paced, recorded lessons
No subscription continuing after programme ends
What sets us apart

Three things we do differently.

The notebook is the lesson

Lessons are not slide decks with a separate exercise file. The notebook contains both the explanation and the working code. You read, run, and modify in the same document. This is how working data scientists actually think through problems.

Code review, not just marking

In the Capstone, instructor reviews at three checkpoints look at your code the way a colleague would in a professional setting — not just checking whether it runs, but whether it is clear, well-structured, and appropriate to the problem.

A panel presentation as the final step

Capstone students present their completed project to a panel of three educators. This is not a formality — it develops the specific skill of explaining your own ML work to people who ask detailed questions about your choices.

Track record

A few numbers.

4
Years running
340+
Students enrolled
6
Active instructors
91%
Programme completion rate
Thailand EdTech Community Recognition
Applied AI education, April 2025
DEPA Thailand Digital Learning Initiative
Participating member, 2024

See how the programmes are structured.

Browse the course details for all three programmes, or get in touch if you want to talk through which one fits where you are right now.

View Programmes Get in Touch