Building AI Systems
That Respect People
A Kuala Lumpur-based team focused on developing recommendation engines, speech recognition systems, and ethical frameworks for organizations that care about implementation quality.
Back to HomeHow Neurova Started
Neurova was founded in 2019 by a group of data scientists and engineers who had spent years working on large-scale AI projects in Singapore and Kuala Lumpur. After seeing countless implementations where the technical capability was strong but the human context was overlooked, the team decided to start something different. The goal was to build AI services that understood the environment they'd be deployed in, not just the algorithms behind them.
The name Neurova combines elements that reflect both neural networks and the idea of renewal. The team wanted a name that acknowledged the technical foundation without sounding like just another tech startup. Based in Kuala Lumpur, the company works with organizations across Southeast Asia and beyond, particularly those in healthcare, logistics, education, and e-commerce sectors where AI can make daily operations more manageable rather than more complicated.
From the beginning, Neurova focused on three areas where careful implementation could make a tangible difference: recommendation systems that help users find what they need without feeling manipulated, speech recognition calibrated for Malaysian language variants and real-world noise conditions, and ethics reviews to help organizations think through the consequences of their AI deployments. These aren't the flashiest offerings in the AI space, but they're the ones where quality of execution matters most.
Today, the team consists of machine learning engineers, linguists, data analysts, and ethics researchers. Projects range from three-week engagements to ongoing partnerships lasting several years. The approach remains consistent: understand the problem thoroughly, build something that works within existing constraints, and leave the client's team equipped to maintain and improve it over time.
The People Behind Neurova
A multidisciplinary team bringing together expertise in machine learning, linguistics, data engineering, and applied ethics.
Dr. Lim Teck Wei
PhD in Machine Learning from NUS. Leads recommendation engine development and oversees model architecture decisions across projects.
Siti Aminah Hassan
Computational linguist specializing in Southeast Asian languages. Manages acoustic model training and multilingual system integration.
Raj Kumar Patel
Background in technology policy and algorithmic fairness. Conducts ethics reviews and stakeholder impact assessments.
How We Work
Quality standards and operational protocols that guide every engagement.
Data Protection Compliance
All projects follow Malaysian Personal Data Protection Act requirements. Data encryption, access controls, and privacy-by-design principles are standard across engagements.
Version Control & Documentation
Every model, configuration change, and deployment is tracked. Documentation is written for the team that will maintain the system, not just for us.
Testing & Validation Protocols
Models undergo bias testing, edge case validation, and performance benchmarking before deployment. A/B testing frameworks ensure measurable improvements.
Stakeholder Communication
Regular progress updates, accessible explanations of technical decisions, and honest assessments of limitations. No overselling capabilities.
Knowledge Transfer
Projects include training sessions for the client's technical team. The goal is to make the system maintainable without ongoing dependence on us.
Ethical Impact Assessment
Every project includes consideration of potential negative consequences. If a proposed implementation raises concerns, we discuss them openly.
What Guides Our Decisions
Neurova operates on the principle that AI systems should make people's work easier without creating new problems. This means being selective about which projects to take on, being transparent about what's technically feasible, and being willing to recommend simpler solutions when they're more appropriate than complex AI implementations.
The team prioritizes projects where there's clear benefit to end users, not just operational efficiency for the client. For recommendation engines, this means focusing on helping people find what they actually need rather than maximizing engagement metrics. For speech recognition, it means optimizing for accuracy in real-world conditions rather than demo performance. For ethics reviews, it means honest assessments even when findings are uncomfortable.
Malaysian context matters in how systems are built. Language variants, cultural norms, regulatory requirements, and infrastructure constraints all influence implementation decisions. A speech recognition system calibrated for Malaysian English performs differently than one trained primarily on American or British accents. These details make the difference between a system that works in practice versus one that only works in controlled conditions.
Interested in Working Together?
We're open to discussing projects where thoughtful AI implementation could make a meaningful difference.
Get in Touch