[ About Cognoscent ]
An independent AI practice built for careful work
Founded in Singapore to serve organisations that treat AI adoption as a considered undertaking, not an item on a quarterly checklist.
Back to HomeFounded with a specific intention
Not a generalist consultancy
Three service lines. No more.
§ Our story
Why Cognoscent exists
Cognoscent was established in Singapore in 2019 by a small group of practitioners who had spent years watching organisations struggle to translate AI interest into anything useful. The common thread was not a shortage of ambition or budget — it was the absence of clear, well-scoped work with honest deliverables.
The name is borrowed from the Latin root for knowing and understanding. It felt right for a practice whose primary commitment is to understand a client's situation accurately before recommending or doing anything. Most of the problems we encounter stem from work that moved before the understanding was there.
We chose to be narrow by design. Rather than offering a broad menu of AI services, we operate in three areas where our team has genuine depth: structured knowledge extraction, reinforcement learning environments, and independent technical due diligence. Each service line has well-defined deliverables, a stated timeline range, and a fixed price. There is no scope expansion without agreement.
Based at Frasers Tower on Cecil Street, we work primarily with enterprise teams in Singapore and the wider Southeast Asia region. We are a small team. We do not take more work than we can deliver well.
Singapore & SEA focus
Small team by design
Understanding first
Every engagement begins with a careful reading of the situation — the data environment, the business context, the constraints. Work that skips this step creates problems downstream. We don't skip it.
Written deliverables
The output of every engagement is a written document — structured, referenced, and designed to remain useful after the engagement closes. We write for decision-makers, not just technical reviewers.
Honest scope
If a client's situation doesn't fit our service lines, we say so directly. We would rather decline an engagement than take work we cannot deliver well. This is not a sales strategy — it is how we operate.
§ The team
The people behind the work
Rohan Krishnamurthy
Founding Principal
Fifteen years working on applied machine learning and knowledge systems across financial services and logistics. Leads the knowledge extraction and due diligence practices.
Selin Tanrıkulu
Principal, RL & Optimisation
Specialises in reinforcement learning environment design and agent training for operational business problems. Background in control systems and simulation engineering.
Li Wei
Analyst, Technical Review
Supports model architecture reviews, data asset assessments, and technical debt analysis across due diligence engagements. Background in NLP and enterprise software.
§ Standards
How we maintain quality across engagements
Confidentiality as standard
Mutual NDAs are signed before any substantive work or data sharing begins. Client information is not referenced in marketing materials or disclosed to third parties.
Documentation of assumptions
Every report includes a section on assumptions and limitations. We consider this as important as the findings themselves — perhaps more so for AI work, where undocumented assumptions are a leading cause of deployment failures.
Structured review process
Each engagement follows a defined internal review process before deliverables are shared. Draft reports are reviewed by a second practitioner independent of the primary engagement team.
No commercial conflicts
We do not receive commissions or referral fees from software vendors, cloud providers, or data platforms. Our recommendations are based on the work — not on commercial relationships.
Data security practices
Client data shared for analysis purposes is stored in encrypted form, access-controlled, and deleted within 30 days of engagement close unless a longer retention period is agreed in writing.
Capacity management
We limit the number of simultaneous engagements to ensure the team is not stretched across too many projects. Quality requires attention, and attention requires space in the schedule.
§ Our expertise
Applied AI work in Singapore and Southeast Asia
Cognoscent operates at the intersection of technical AI work and organisational decision-making. The three service lines reflect different points where careful, structured work has a clear impact: extracting and formalising institutional knowledge, designing learning environments for operational AI systems, and providing objective assessments of AI assets under evaluation.
Singapore's position as a regional technology hub makes it a natural base for this kind of work. Enterprise teams at major corporations, investment groups evaluating technology assets, and government-linked entities building AI capabilities — all operate within a few kilometres of our Cecil Street office. We understand the business context, the regulatory environment, and the particular challenges of AI adoption in this region.
The practice draws on experience in natural language processing, knowledge representation, simulation and control systems, and the technical evaluation of machine learning pipelines. We are not generalists who have recently expanded into AI — the team has worked in this space for over a decade, through the cycles of enthusiasm and disappointment that have preceded the current period of serious adoption.
Ready to discuss your AI challenge?
Describe your situation and we'll respond with an honest view of whether we can help.
Start the Conversation