Participant Feedback
What developers say after going through the courses.
Feedback from people who completed the Foundations and LLM Engineering programmes, and from teams that ran the Team Capability Programme.
Back to Home4+
Years running courses
240+
Individual participants
4.6/5
Average course rating
18
Teams completed the business programme
Reviews
From participants across our cohorts
Reviews are from people who completed the courses. We have not filtered by rating.
Farid Hassan
Backend developer · Kuala Lumpur
The Foundations course does not pretend you already understand calculus in the context of training loops. That was the thing I needed most — an explanation of why the maths matters for the code I was writing, not a pure maths lecture. The assignments were hard in the right way; two of them took me longer than the estimate. Tutor feedback was specific and pushed me to think about regularisation differently.
ML Foundations · June 2025
Priya Letchumanan
Data analyst · Petaling Jaya
I had tried two other online ML courses before this one and stopped both partway through because I could not tell whether I was actually learning or just clicking through videos. The live sessions here were different — you can ask a question and get a direct answer, not a forum thread that someone maybe answers later. The twelve weeks felt like a lot going in, but by week eight I was glad the pacing was not faster.
ML Foundations · June 2025
Zulaikha Ahmad
Software engineer · Shah Alam
The LLM Engineering course is the first one I've taken that covered evaluation properly — not just "does the output look right" but how to write regression tests and catch regressions before they go to production. The project defence session was stressful but useful; having to explain your architecture out loud to someone who knows the field identifies gaps that no automated test would catch.
LLM Engineering · May 2025
Kamarul Mukhtar
Platform engineer · Cyberjaya
I appreciated that the course page listed the workload in hours before it listed the price. That is a small thing but it meant I went in with an honest expectation rather than finding out week three that six hours is more like nine if you engage with the exercises properly. One feedback: I would have liked more time on the retrieval indexing section — we moved through it faster than felt comfortable. But the rest of the scope was well-paced.
LLM Engineering · May 2025
Nurul Cahaya
Junior data scientist · Kuala Lumpur
What stayed with me from the Foundations course is how the tutor explained the relationship between the loss function shape and why certain optimisers behave the way they do on particular datasets. That section took a full week and I thought it was going to be too much detail — it was not. It made the rest of the course make more sense.
ML Foundations · July 2025
Rajesh Subramaniam
Engineering manager · Bangsar South
We ran the Team Capability Programme for ten engineers. The capability review before the programme started was more thorough than I expected — they asked specific questions about our deployment stack and what kinds of internal projects we had in the backlog. The syllabus that came out of it was noticeably different from the published individual course outline, which is exactly what we needed. The follow-up clinics at month three were useful for questions that only arose once people were applying the learning to live work.
Team Capability Programme · April 2025
Case Studies
Three teams, three different starting points
Fintech engineering team, Kuala Lumpur — 8 participants
Team Capability Programme · 16 weeks · Completed May 2025
Challenge
The team had been integrating third-party ML APIs into their risk scoring pipeline but had limited understanding of what was happening inside those models. When an API change produced unexpected outputs, the team could not reason about why or how to test for similar issues in future.
Approach
The capability review identified the gap as conceptual rather than tooling-based. The programme prioritised evaluation methodology and failure mode analysis over new model training. Code review clinics ran against the team's existing risk pipeline, not sample code. The internal project was a monitoring layer for the existing API integration.
Outcome
By the end of the programme the team had a working evaluation harness for their API integration and had documented the expected input distribution for their three main use cases. The follow-up clinic at month three addressed one regression that occurred after an upstream model update.
"The code review sessions were the most immediately useful part. Having someone read our actual code rather than example code made the feedback directly applicable." — Engineering Lead
Individual developer, LLM Engineering course
Applied Language Model Engineering · 10 weeks · Completed June 2025
Background
A developer with three years of Python experience and completion of a prior ML course, joining to learn how to build and deploy a RAG system for an internal document search tool at their company.
Experience
The course covered retrieval-augmented generation in weeks five and six. The participant spent approximately twelve hours that week rather than the stated ten, particularly on the indexing and chunk-size tuning sections. The project was a working document search API connected to a private document store.
Assessment
The project was deployed to the sandbox environment and presented in the fifteen-minute defence session. The tutor's review focused on latency budgeting choices and the evaluation strategy for retrieval quality. The participant had not fully documented the latency tradeoffs and revised the approach after the session.
"The defence session was uncomfortable in exactly the right way. Explaining your choices out loud is a better test than getting a passing grade on a quiz." — Participant
Logistics company engineering team, Selangor — 12 participants
Team Capability Programme · 16 weeks · Completed April 2025
Challenge
The team wanted to move route optimisation from a third-party SaaS tool to an in-house model they could retrain as their delivery network changed. No team member had prior experience training production ML models.
Approach
The capability review found mixed Python proficiency across the team. The programme ran two tracks in parallel for the first four weeks — a foundations track for five team members and a more advanced track for the remaining seven. Both tracks converged on the shared internal project in week five.
Outcome
The internal project produced a working prototype route optimisation model, trained on the team's historical delivery data, with a documented evaluation process. The written curriculum handover included the syllabus, session recordings, and the internal project specification. The six-month follow-up clinic discussed a retraining pipeline they had built independently after the programme ended.
"Having the recorded sessions meant we could bring a new team member up to speed six months later without starting from scratch." — CTO
Get in Touch
Questions before you decide?
We answer questions about prerequisites, workload, cohort dates, and team programme scope directly — no sales process, no follow-up call unless you ask for one.
-
Phone
+60 3 2794 1568 -
Email
[email protected] -
Office Hours
Mon–Fri 9:00 AM – 6:00 PM MYT
Sat 10:00 AM – 2:00 PM MYT -
Address
45 Jalan Pinang, 50450 Kuala Lumpur
Read the syllabus, then ask your question.
The course details are on the Solutions page. Once you have read them, the enquiry form is a direct line to us.