§ 01 — About the Practice
Practised engineering applied to AI problems in Malaysian organisations
Anggun Systems was established to fill a particular gap in the Malaysian market: the gap between organisations that understood AI had something to offer them, and organisations that had AI systems that ran reliably in production.
Return to Homepage§ 02 — Our Story
How Anggun Systems came to be
This section describes the circumstances that led to the founding of Anggun Systems, the problem we set out to address, and the working principles we formed in response. It does not describe capabilities we do not have.
Anggun Systems was founded in Kuala Lumpur by a small team of engineers who had spent several years working on data and software projects for Malaysian firms across the financial services, manufacturing, and professional services sectors. The observation that prompted the company's formation was consistent across those engagements: organisations were encountering AI systems that were either unsuited to their specific problem, undocumented, or not properly evaluated before deployment.
The consequences were predictable. Staff could not assess when to trust the system's output and when to override it. Leadership could not explain the system's behaviour to auditors or regulators. When the team that built the system moved on, the system became a black box that no one could maintain. These are not exotic failure modes; they appear in any field when engineering discipline is not applied consistently.
We formed Anggun Systems around a simpler proposition: AI systems, like any other software, should be scoped before they are built, tested before they are deployed, and documented so that the people who use them and the people who come after them can understand what the system does and where its limits are. This approach is less exciting than the claims made about AI in general, but it is considerably more useful to an organisation that must rely on the software in practice.
Our three service lines — custom model development, internal knowledge retrieval, and AI policy drafting — were chosen because they represent the three most common, tractable problems we observed. Each can be completed in a defined timeframe, delivered with documentation, and handed over to the client without creating a dependency on our continued involvement.
We operate from our office on Jalan Tunku Abdul Rahman, serve clients across Peninsular Malaysia, and conduct engagements in English. We are a small practice by design; every engagement involves senior engineers, not junior staff under indirect supervision.
§ 03 — The Team
The people who conduct the work
Anggun Systems is a small practice. The individuals described here are directly involved in client engagements. We do not use subcontractors without client agreement.
Zainal Rashid
Principal Engineer
Zainal leads model development engagements and is responsible for the technical evaluation methodology used across all three service lines. He previously worked in applied machine learning at a Kuala Lumpur-based fintech for seven years.
Nurul Kamariah
Systems Engineer
Nurul designs and builds the retrieval-augmented systems that underpin our knowledge retrieval service. She has a background in information architecture and enterprise search, and holds a Computer Science degree from Universiti Malaya.
Faris Halim
Policy Analyst
Faris leads the AI policy drafting engagements. His background is in technology law and regulatory affairs, with specific experience in the Malaysian Personal Data Protection Act and sector-level guidance from Bank Negara and the Securities Commission.
§ 04 — Working Standards
Standards we apply to every engagement
These are not aspirational values; they are the concrete practices we follow on every project. Where a project constraint prevents us from applying one of these standards, we flag this to the client at scoping, before work begins.
Data handling agreement
A written data handling agreement is signed before any client data is shared. Data is used solely for the stated engagement and is not retained beyond project completion.
Performance evaluation
Every system we deliver is tested against a defined evaluation methodology before handover. Results are shared with the client in writing, including failure cases and known limitations.
Handover documentation
All deliverables include documentation written for a successor team — engineers who were not involved in the build and will need to maintain, extend, or audit the system.
PDPA compliance
Our data practices are designed to operate within the Malaysian Personal Data Protection Act. Where a client's sector carries additional obligations, we work within those as well.
Agreed-in-writing timelines
Engagement scope and timeline are agreed in writing before work begins. We do not alter the scope without a written change request that restates the timeline and any price adjustment.
Senior engineer involvement
Engagements are conducted by the named engineers who lead our three service lines. We do not delegate client-facing work to junior staff without explicit agreement.
§ 05 — Our Position in the Market
AI engineering for organisations that need the work done properly
The AI consulting market contains a wide range of practitioners, from large technology firms offering platform-level AI integration to individual freelancers working across whatever tools are currently available. Anggun Systems occupies a specific position within that market: a small practice focused on three well-defined service lines, conducted with engineering discipline, and delivered with documentation.
We work with Malaysian organisations across sectors including professional services, financial services, manufacturing, logistics, and healthcare administration. Our clients share a common characteristic: they have a specific, bounded problem that off-the-shelf AI tools address poorly, and they have an internal requirement — whether regulatory, operational, or risk-related — that the system they deploy be documented and evaluable.
We are not a research organisation. We do not publish papers, develop novel architectures, or participate in the academic AI literature. We apply existing, proven methods to client problems and deliver systems that run. The distinction matters because it shapes how we scope engagements, what we commit to, and what we honestly decline to do.
Our presence in Kuala Lumpur is deliberate. Understanding the regulatory environment, the business culture, and the practical constraints that Malaysian firms operate under is relevant to the quality of the work. Remote-only AI consultancy from outside Malaysia is possible, but it carries costs that are not always visible at the outset of an engagement.
§ 06 — Begin a Conversation
Speak with us about your situation
The first conversation is a scoping discussion. We will tell you whether your problem fits one of our service lines, and if it does not, we will say so directly.
Get in Touch