What Learners
Say After Completing Tracks
These accounts are from people who completed at least one Tensora track. They describe what the work involved, what they found difficult, and how it affected what they do next.
Back to Home180+
Learners Completed
4.7
Average Rating
3
Track Options
92%
Cohort Completion
From Past Cohorts
Pichaya Teerasombat
Data Analyst · Bangkok
"I had used scikit-learn before but mostly by copying tutorials. The Applied ML track made me think through why I was making specific preprocessing choices rather than just running the same pipeline on every dataset. The feedback on my final submission was genuinely useful — not just 'this looks good'."
Nathamon Laohaphan
Software Developer · Chiang Mai
"The Data Engineering track filled in gaps I did not know I had. I write application code for a living, but had never thought carefully about how data moves before it reaches a model. Some modules were more demanding than I expected for the time commitment, but the scope per week is at least honest about that upfront."
Warut Chantarasak
ML Engineer · Chiang Mai
"I had been training models for about two years and had never shipped one properly. The MLOps track gave me the vocabulary and the working patterns to actually get something into production. The monitoring module in particular — I had ignored that part completely in my own projects."
Suphansa Poomsiri
Research Assistant · Chiang Mai
"I am an academic, not an engineer, and I was not sure the Applied ML track would suit someone from a statistics background rather than software. It did — the first module worked from the dataset up rather than from a framework down, which matched how I think about problems."
Kritsanaphong Jiraporn
Backend Developer · Bangkok
"Data Engineering Foundations covered things I genuinely did not know — the module on validation and quality checks in particular. I wish the cohort had a bit more synchronous time for discussion, but the async feedback worked well enough. I completed this while working full time, which would not have been possible in a more intensive format."
Areeya Nontapat
DevOps Engineer · Chiang Mai
"I came from infrastructure work and picked up the MLOps track to understand the model side of deployments I was already helping manage. The track bridged that gap well — I now understand what the data scientists I work with are actually asking for when they describe serving requirements."
Learner Journeys in Detail
Thanakorn Wattanasin — Applied ML Track
Marketing Analytics Manager · Bangkok · Completed May 2025
Starting Point
Thanakorn ran A/B tests and built dashboards but had not used predictive modelling in his work. He wanted to understand whether models could meaningfully improve campaign targeting, but did not know how to approach building one.
What the Track Covered
Over seven weeks, Thanakorn worked through dataset preparation, feature construction from behavioural data, and training a classification model to score leads by likelihood to convert. He documented each decision in a structured notebook.
Outcome
He applied a simplified version of the model in a pilot campaign, which reduced spend on low-probability contacts by around 18%. He also came out with a notebook he could share with colleagues to explain the logic.
"The track gave me a working model and, more usefully, an understanding of the choices I made to get there — which I could explain to people who were not in the cohort."
Phantipha Charuwan — MLOps & Deployment Track
ML Researcher · Chiang Mai · Completed June 2025
Starting Point
Phantipha had spent two years training NLP models for her research group, but every model lived in a notebook. When colleagues asked to use outputs in another system, she had no clear process for that handoff.
What the Track Covered
The MLOps track walked her through packaging a trained model, building a simple inference endpoint, adding logging and basic drift monitoring, and documenting the full setup so it could be reproduced without her present.
Outcome
She deployed a working inference API for one of her research models, which a colleague integrated into a separate system within two weeks. The monitoring setup she built has flagged one significant input distribution shift since deployment.
"I went from models that only I could run to a model that other people could actually use. That was the gap the track filled."
Questions Before Enrolling?
Mon–Fri 09:00–18:00
We respond within one working day
55 Nimmanhaemin Rd
Chiang Mai 50200
In-person or remote
See Whether a Track Fits Where You Are
The accounts above are from people who came in with different backgrounds and different goals. If you are unsure how one of the tracks maps to your situation, get in touch and we can work that out with you.
Get in Touch