Our Story
How Neural Loft came to be, and what it stands for.
Neural Loft began in 2021 when a small group of machine learning practitioners based in Selangor found themselves repeatedly asked the same question by curious colleagues and friends: how do you actually learn this material? Not which platform, but how — at what pace, with what kind of attention, and with whom.
The answer they kept returning to was the same. You read carefully. You write code by hand and look closely at what it does. You discuss what confuses you. You allow yourself to not understand something yet, and you return to it. None of this required a new kind of curriculum. It required the conditions for a certain kind of study — a smaller room, a slower pace, and someone who would give honest feedback on your work.
Neural Loft was set up to provide those conditions online, with three programmes designed around what a working learner in Malaysia actually needs: substance, clarity, and a school that is prepared to be patient.
Our Mission
To offer AI and machine learning programmes that ask something genuine of the learner — and give something substantial in return. We do not race through a syllabus. We do not offer certificates in place of understanding. We design each course to be a proper encounter with the material.
Our Values
- Honesty about difficulty. We describe what each programme asks of a learner without minimising the effort involved.
- Substance over speed. Ten weeks is enough time to understand a topic well if used carefully. We use it carefully.
- Respect for the learner. Enquiries are answered with genuine attention. Sample materials are shared without conditions.
The People
Those who teach and guide at Neural Loft
Rajan Arumugam
Lead Mentor — Data Programming
A practitioner with twelve years in data engineering and applied Python development. Rajan designs the Data-Oriented Programming curriculum and leads its weekly sessions, with particular attention to how learners develop confidence when reading unfamiliar code.
Lim Hui Shan
Lead Mentor — NLP Programme
Hui Shan holds a postgraduate qualification in computational linguistics and has worked in applied NLP roles across Singapore and Malaysia. She leads the NLP programme, focusing on principled model evaluation and the discipline of reading results honestly rather than selectively.
Zulaikha Ibrahim
Study Group Facilitator
Zulaikha facilitates the research paper reading group and has been working in ML research since 2018. She selects each quarter's papers with attention to their clarity and relevance, and runs sessions in a way that encourages scepticism over compliance.
Standards
How we approach quality in teaching and operations
Curriculum Review
Each programme is reviewed by its lead mentor at the end of every cohort. Exercises, written materials, and session structures are updated to reflect what learners found unclear or underexplained.
Cohort Sizes
Groups are kept deliberately small — typically six to ten participants — so that each learner receives genuine attention during sessions and personalised feedback on submitted work.
Data Privacy
Learner data is held securely, used only for course administration, and never shared with third parties for commercial purposes. Requests to access or remove personal data are handled within five working days.
Pacing Commitments
If a learner misses a session due to illness or work, recordings are made available and the mentor will provide a written summary of what was covered and how to catch up without confusion.
Honest Admissions
We do not admit learners to a programme we believe they are not ready for. Where a different starting point would serve better, we say so plainly and suggest a suitable path forward.
Feedback Culture
At the end of each programme, participants complete a detailed feedback form. Responses are read by the lead mentor and used to inform the next cohort's structure. Negative feedback is treated as useful information, not a problem to manage.
Our Approach
AI education in Malaysia, built around the practice of study
The field of machine learning is wide, and the literature moves quickly. Neural Loft does not try to keep pace with every development. Instead, each programme is designed to build the kind of grounded technical understanding that a learner can use to navigate new material independently — understanding well enough to know what questions to ask and where to look for answers.
Subang Jaya offers a practical and well-connected base for learners across the Klang Valley. Our online delivery means that working professionals in Petaling Jaya, Shah Alam, Kuala Lumpur, and further afield can join a programme without managing commute time. Sessions are scheduled with working hours in mind and the written materials are designed to be studied during the week at a learner's own pace.
We are a small school by design. Large cohorts are not part of our model. The three programmes we currently offer are the ones we can teach well, and we have no plans to expand that list until we are satisfied each existing programme is working as it should.
Next step
Read the programme details and ask us anything.
The enquiry form is always open. We reply within two working days with honest information about which programme might suit where you are now.
Send an Enquiry