What People Say After Finishing a Track

Challenge

Working in logistics operations, Phonchai was handling large spreadsheets manually. He knew data tools existed but had never written a line of code and was not sure where to start without wasting time on the wrong material.

Approach

He enrolled in the Data & Code Foundations track. Working roughly eight hours a week around shift patterns, he moved through the Python and pandas modules steadily. When he submitted his first exercise, the mentor's notes pointed out a loop he was writing manually that pandas handles natively.

Outcome

By week nine, Phonchai had completed a data pipeline project that cleaned, transformed, and summarised delivery records automatically. A task that previously took three hours of manual work took under two minutes. He is now considering the Applied Models Studio track.

Challenge

As a product manager working closely with data teams, Siriporn understood what models were supposed to do but not how they were built. She had basic Python from a previous short course but had never built a model from scratch.

Approach

She enrolled in Applied Models Studio and worked through the feature engineering section twice — the mentor's feedback on her first submission flagged data leakage she had not spotted. That note alone changed how she thought about the split between training and test data.

Outcome

Her portfolio project — a churn prediction model for a hypothetical subscription product — is something she can now walk through technically with the engineers on her team. She says conversations about modelling decisions are different now that she understands what the choices actually mean.

Challenge

Wanchai could build models that worked well in notebooks but struggled when it came to how they would be deployed, monitored, and updated in production. His existing team had no agreed architecture and things broke frequently after handoff.

Approach

The Scalable AI Track's system design modules gave him a framework for thinking about components and their interfaces. The peer collaboration sections — where he and two other learners were each given a different constraint to design around — turned abstract patterns into real decisions.

Outcome

He brought an architecture proposal to his team at week sixteen and it was adopted with adjustments. His capstone — a documented AI serving architecture with monitoring and rollback planning — gave the whole track a concrete reference to close with. He finished in 19 weeks.