What students say about the programmes.

I had tried two other Python courses online before this one and kept giving up around week three. The difference here was the notebook exercises — they were not just drill, they were small analysis tasks with actual data. Getting written comments back on my submitted notebooks helped me understand where my reasoning was off, not just where my code was wrong. I finished the six weeks.

The ML Pathway is a serious commitment. Fourteen weeks at six to eight hours a week is real — I want to say that upfront. But the mentor office hours were worth it. I could bring a specific confusion — why my model was not generalising, for example — and actually work through it with someone rather than spending a weekend reading stack overflow. I have three projects in my GitHub now that I can talk about.

I finished the Capstone in February. The three code reviews were what I was not expecting — not just pass or fail comments, but specific notes on my choices. Why I had designed the architecture the way I did, whether the data augmentation strategy made sense, what the evaluation metrics were actually telling me. The panel presentation felt difficult at the time but I came out of it knowing my own project far better.

I work as an analyst and already used Excel well, but I wanted to move into Python for larger datasets. Data Foundations was the right level. A few of the early weeks were slower than I needed, but I appreciated that the notebooks were cumulative — week 5 used what I had built in week 2. The written feedback on my second notebook pointed out a bias in how I had handled missing values that I would have carried forward.

I took the ML Pathway from Phuket while working full-time in hospitality IT. Being able to do the lessons at whatever time worked for me was not a small thing — I often studied after midnight. The three portfolio projects gave me something to show in interviews. I am now in a data analyst role that uses the skills directly.

The Capstone is honestly intense. I hit a wall around week nine when my model was not converging and I was not sure whether the problem was the architecture, the data, or how I had set up training. The mentor review at that checkpoint helped me isolate it. I would say expect to spend more time on it than you think, and use the mentor access while you have it.