Building Trust in Artificial Intelligence
We started prismars.pro because we believed AI quality should never be an afterthought. It should be embedded in the way teams build, ship, and maintain intelligent systems.
Back to HomeWho We Are
prismars.pro was founded in Kuala Lumpur by a group of engineers and data scientists who had spent years building and deploying AI systems across Southeast Asia. Through that experience, one pattern stood out again and again: teams were investing heavily in model development, but far less in validating that those models worked reliably once they left the lab environment.
That gap — between what a model could do in controlled conditions and what it actually did in production — became the core problem we set out to address. We designed our practice around the idea that AI systems deserve the same rigour of testing and monitoring that critical software infrastructure receives.
Today, prismars.pro works with organisations across Malaysia and the wider ASEAN region, helping them establish structured quality assurance, stress-test their models against adversarial conditions, and build observability into their AI operations. We are not a consultancy that hands over a slide deck. We work alongside your engineering teams, producing real test suites, real dashboards, and real operational runbooks.
Our Mission
To make AI quality measurable, manageable, and part of every organisation's standard engineering practice — not an optional extra, but a foundational habit.
The Team Behind prismars.pro
Arif Khalid
Lead AI Quality Engineer
Eight years of experience in machine learning engineering and software quality. Previously led QA for an AI-driven logistics platform serving Southeast Asian markets.
Su Lin Ong
Adversarial Testing Specialist
Background in cybersecurity and applied machine learning. Focuses on building robust evaluation suites that expose model weaknesses before they surface in production.
Rajesh Nair
Monitoring & Observability Lead
Ten years in DevOps and platform engineering, now specialising in production ML observability. Experienced in designing alerting systems for high-throughput AI services.
Quality and Safety Protocols
Every engagement follows documented protocols designed to maintain consistency, reproducibility, and professional integrity.
Data Confidentiality
All client data is handled under strict NDA agreements. We support on-premise engagements where data governance policies require zero external transfer.
Documented Methodology
Our testing processes follow repeatable, version-controlled methodologies. Every test case, finding, and recommendation is traceable back to its source.
Continuous Learning
Our team stays current with developments in AI safety research, adversarial ML, and monitoring best practices through ongoing training and community engagement.
Reproducible Testing
Test environments are containerised and versioned, allowing any evaluation to be re-run under identical conditions for verification or audit purposes.
Ethical AI Practices
We test for fairness, bias, and unintended discrimination as part of our standard evaluation suite — because quality includes responsible behaviour.
Regulatory Awareness
We track emerging AI governance frameworks in Malaysia and ASEAN, helping clients prepare for compliance requirements before they become mandatory.
What Drives Our Work
AI systems are increasingly central to business operations across Malaysia — from financial services and healthcare to logistics and e-commerce. Yet the quality practices surrounding these systems often lag behind traditional software engineering standards. At prismars.pro, we see that as a solvable challenge, not an inevitable one.
Our approach is grounded in the belief that testing and validation should be systematic, not sporadic. When an organisation can point to specific metrics — mean-time-to-detect for drift, regression rates across model versions, robustness scores under adversarial conditions — they move from hoping their AI works to knowing where it stands.
We also believe strongly in knowledge transfer. Our engagements are designed so that when we leave, your team has the tools, documentation, and understanding to continue the quality practices we established together. That is the measure of a successful partnership — not dependency, but capability.
For organisations across Kuala Lumpur and the broader ASEAN technology landscape, prismars.pro offers a practical, structured path to AI systems that perform as expected — in the lab, in staging, and critically, in production. We focus on building lasting quality culture, one engagement at a time.
Interested in Working Together?
We would be glad to hear about the AI challenges your organisation is facing. Reach out for an initial conversation — no pressure, just an honest assessment of how we might help.
Contact Our Team