A Small Team with a Clear Focus:
AI Compute Energy, Understood
Vantsuro was founded to give platform and facilities teams a calm, factual view of what their AI infrastructure consumes — and what that means for their operations.
How Vantsuro Came to Exist
Vantsuro grew out of conversations that kept happening at the edge of AI infrastructure projects in Malaysia. Engineers and facilities managers were building and expanding GPU clusters with real pace, but the question of how much energy those clusters actually used — and why — kept being deferred. It was treated as someone else's problem.
We noticed that the available information was either highly technical research with limited practical application, or vendor materials with commercial angles. There wasn't much in between: structured, plain-language support for the people managing the infrastructure day-to-day.
We set up Vantsuro to sit in that space. Based at Technology Park Malaysia in Kuala Lumpur, we work directly with platform engineers, data centre managers, and IT leads at organisations building out AI compute capacity across the region.
What We Are Here to Do
Our work is to share information clearly and help teams develop their own steady, documented understanding of compute energy use. We do not make environmental claims on behalf of clients, and we do not engage in regulated certification processes.
Every engagement we deliver is designed around one principle: respect for the team's own judgment. We present figures, explain how they were arrived at, and offer practical frameworks for tracking and discussing them internally. What an organisation does with that information is their decision.
"We share information and support steady practice — without preaching and without an agenda beyond the engagement itself."
— Vantsuro working principles
People Behind Vantsuro
A focused team with backgrounds spanning infrastructure engineering, data centre operations, and technical communication.
Razlan Hamid
Principal Advisor
Infrastructure engineer with twelve years working on GPU and HPC deployments across Malaysia and Singapore. Leads review and advisory engagements.
Suraya Nordin
Platform & Workload Specialist
Specialises in workload scheduling and utilisation analysis for large-scale compute. Designs workshop content and practices checklists.
Chen Wei Liang
Technical Documentation Lead
Responsible for written deliverables and handbook production. Ensures technical figures are presented clearly for non-specialist stakeholders.
Our Working Standards
The principles that shape how we prepare, deliver, and follow up on every engagement.
Documented Estimation Methods
We use published and peer-reviewed sources for estimation. Where figures are approximated, we say so clearly and cite the method used.
Client Data Confidentiality
All infrastructure information, utilisation data, and operational details shared during an engagement are held in confidence and not disclosed to any third party.
Vendor Independence
We hold no commercial arrangements with hardware vendors, cloud providers, or software suppliers. Guidance reflects operational facts, not referral incentives.
Written Deliverables Every Time
Every engagement ends with a written output. We do not treat verbal-only sessions as complete. Documentation is part of the service, not an optional add-on.
Scope Clarity Before We Start
Before any engagement begins, we confirm in writing what is included and what is outside scope. This protects both parties and ensures expectations are aligned.
Plain Language Throughout
We write and speak plainly. Technical terms are explained when used. Reports are written so that facilities managers and finance leads can read them alongside engineers.
AI Compute Energy: A Practical Field
The energy use of AI infrastructure has become a meaningful operational consideration for many organisations in Malaysia and across Southeast Asia. As GPU clusters scale — whether in private data centres, colocation facilities, or cloud environments — the costs and operational implications of power consumption require direct attention from the people who manage those systems.
At Vantsuro, our work sits at the intersection of infrastructure knowledge and clear communication. We understand how GPU power draw, cooling overhead, and utilisation patterns interact, and we know how to present that understanding in a form that is useful to the people responsible for making decisions about it.
Our three service offerings address this at different stages: an initial review for teams wanting a baseline picture, a workshop for teams ready to explore operational adjustments, and a longer advisory engagement for organisations building internal processes. Each is designed to produce something usable, not just a conversation.
We work with teams at Technology Park Malaysia and across the Klang Valley, and we deliver remotely for clients further afield in Malaysia. Our approach does not depend on expensive tooling or proprietary platforms — it relies on structured thinking, clear documentation, and direct engagement with the people who know the infrastructure best.
Want to Know More About Working With Us?
We're happy to have a brief conversation before any engagement is agreed. Reach out to describe your setup and what you're trying to understand.
Get in Touch