The practice behind the work
Dataveil is a Singapore-based AI advisory practice. We work with organisations that are navigating real questions about AI — not headlines.
Back to HomeWhere Dataveil came from
Dataveil was founded out of a straightforward observation: most organisations working on AI projects spend enormous effort on model development, and comparatively little on the conditions those models need to function responsibly at scale — privacy boundaries, production performance, and the decision-making layers that sit above raw outputs.
The practice was established in Singapore to work directly with organisations across financial services, healthcare, logistics, and the public sector. Singapore's regulatory environment — particularly PDPA requirements and MAS guidance on technology risk — creates a distinctive set of constraints that require careful navigation. That context shaped how Dataveil's engagements are structured.
Rather than broad consulting arrangements, Dataveil works in focused, time-bounded engagements with defined deliverables. Each engagement ends with documentation, configurations, and knowledge transfer that remain with the client. The intent is that clients leave more capable than when they started.
What we are here to do
Help organisations make considered decisions about AI — and implement them in ways that hold up under production conditions, regulatory scrutiny, and the practical complexity of real data environments.
How we work
- Honest scoping — we decline work that isn't a genuine fit
- Structured knowledge transfer at the end of every engagement
- No padding engagements — scope is sized to what is actually needed
- Clear communication when timelines or scope shift
The people you work with
A small team with deep specialisation, rather than a large firm with generalist coverage.
Ethan Lim
Lead AI Architect
Works across model design and privacy-preserving AI architecture. Spent eight years in financial technology before moving into advisory work. Holds an MSc in Computer Science from NUS.
Shreya Rajan
Data Privacy Lead
Specialises in regulatory compliance for AI systems, including PDPA, GDPR, and MAS technology risk frameworks. Background in legal technology and data governance across Singapore and the UK.
Wei Chen
MLOps & Inference Engineer
Focuses on production AI infrastructure — latency reduction, cost optimisation, and deployment reliability. Previously led inference engineering at a regional e-commerce platform.
How quality is maintained
Every engagement follows a consistent set of practices — adapted to the service type but never shortened.
PDPA & GDPR Alignment
All engagements are reviewed for compliance with Singapore's Personal Data Protection Act and, where applicable, the EU General Data Protection Regulation. Data handling is documented throughout.
Structured Deliverable Review
Every deliverable passes through a structured internal review before being shared with clients. Technical documents are reviewed for accuracy; recommendations are reviewed for feasibility in context.
Confidentiality by Default
All engagement materials are treated as confidential. NDAs are standard. Client data and internal documentation are never shared across engagements or referenced in external communications.
MAS Technology Risk Awareness
For clients in financial services, our guidance is reviewed against the Monetary Authority of Singapore's Technology Risk Management Guidelines to ensure alignment with sector expectations.
Benchmarked Performance Claims
Any performance improvements referenced in deliverables are supported by before-and-after benchmarking data. We do not make claims about speed or efficiency gains without measurement to back them.
Knowledge Transfer Protocol
Every engagement closes with a structured handover session, written documentation in plain language, and — where relevant — training sessions to ensure your team can maintain and adapt the work independently.
AI advisory for complex operating environments
Singapore's position as a regional hub for financial services, logistics, healthcare administration, and digital infrastructure means that organisations operating here face a distinctive combination of pressures when it comes to AI adoption. Regulatory frameworks are active and evolving. Data volumes are significant. Stakeholder expectations — from boards, from regulators, from partners — are increasing. At the same time, internal AI teams are often small relative to the scale of the challenges they are asked to address.
Dataveil works within this context. Our engagements are designed for organisations that have moved past initial AI exploration and are now dealing with real-world friction — data that doesn't behave as expected, models that work in staging but underperform in production, privacy requirements that weren't fully considered at the design stage, or business intelligence environments that have grown complex enough to obscure rather than illuminate.
The three service areas — privacy-preserving AI consultation, inference optimisation, and AI-integrated business intelligence — each address a distinct layer of this operating environment. Together they cover the design, performance, and decision-support dimensions of putting AI to work at scale. Engagements can be taken individually based on where the most acute need sits, or sequenced as an organisation's AI capabilities mature.
What Dataveil does not do is provide coverage for its own sake. Engagements are scoped honestly, and if the right answer is a smaller engagement — or a referral to a different specialist — that is the recommendation made. The practice's reputation is built on the quality of what is delivered and the utility of what clients take away, not on the breadth of services promoted.
Want to understand whether there's a good fit?
A brief conversation about your situation is the best starting point. No commitment required on either side.
Get in Touch