OUR COMPANY
A firm that thinks before it builds
We started Lambent because the AI conversation was moving too fast for most organisations to keep up — and too slowly on what actually matters.
← Back to HomeOUR STORY
Where Lambent began
Lambent was founded in Wan Chai in early 2022 by a small group of practitioners who had spent years working at the intersection of machine learning research and enterprise software. We had watched the same patterns repeat across projects: organisations adopting AI tools before understanding what problem they were solving, or waiting for permission from a conference stage that never came.
We built Lambent to fill a quieter role. Not to sell platforms, not to chase press releases, but to sit with organisations and help them think clearly about where AI can genuinely contribute — and where it adds complexity without value.
Our name comes from the quality of light that seems to glow from within an object rather than bouncing off its surface. That is what we are after in our work: insight that comes from understanding, not from surface-level demonstration.
Today we maintain a small team in Hong Kong, working with a mix of financial services firms, professional services organisations, and technology companies across the region.
Our Mission
To help organisations in Hong Kong and the wider Asia-Pacific region introduce AI capabilities that are well-understood by the people who use them, proportionate to the problems they address, and maintained to remain accurate over time.
Our Values
- Clarity — we prefer honest, direct communication over impressive-sounding language.
- Proportionality — the right tool for the right problem, not the most sophisticated available.
- Continuity — we stay involved through maintenance, not just initial delivery.
OUR TEAM
The people behind the work
Marcus Kwong
Co-founder & Principal
Marcus leads client engagements and oversees technical delivery. He spent eight years in NLP research before moving into applied AI consulting across the financial sector.
Sophie Lam
Co-founder & Strategy
Sophie focuses on executive communication and AI strategy. Before Lambent she led digital transformation programmes for professional services firms across APAC.
Rajan Nair
ML Engineer
Rajan handles the implementation side — embedding model integration, retraining pipelines, and infrastructure. He brings a preference for simple, auditable systems.
HOW WE WORK
Standards we hold ourselves to
Written scope for every engagement
We document what we will do, what we will not do, and how success will be measured — before any work begins.
Data handling agreements
All client data is handled under a written agreement specifying access, retention, and deletion — in line with the Personal Data (Privacy) Ordinance (HK).
Version control on all models
Every model we retrain or deploy is versioned. You can always roll back to a previous version if needed, without loss of work.
Performance evaluation at each stage
We evaluate model performance before and after retraining, sharing results in plain language alongside the technical metrics.
Knowledge transfer included
We walk your team through how the systems we build actually work — so your people can manage, question, and maintain what we deliver.
No retainer lock-in
We do not require ongoing retainer commitments. Each project stands on its own. Continued engagement is by choice, not by contract.
OUR EXPERTISE
AI services grounded in practice
Lambent draws on applied machine learning experience across natural language processing, information retrieval, and production model lifecycle management. Our work spans multiple sectors in the Hong Kong and broader APAC market, giving us familiarity with the regulatory environment, data governance expectations, and organisational culture that shapes how AI can realistically be used here.
We do not provide generic AI consulting. Each engagement addresses a specific operational need — whether that is improving how staff find information internally, extending the useful life of a deployed classification or recommendation model, or equipping senior decision-makers with a clear picture of where AI investment makes sense for their organisation.
Our approach favours smaller, well-scoped implementations over large transformation programmes. AI delivered at the right scale, with proper understanding from the people who depend on it, tends to produce more durable results than ambitious projects that outpace internal capacity.
Interested in working with us?
A brief conversation is a good place to start. We are happy to discuss your situation before any commitment is made.
Get in Touch