Helping Teams Think
Clearly About Compute
Wexar was built around a straightforward observation: most AI workload planning struggles not from a lack of data, but from a lack of structure. We build tools that give teams the structure they need.
Back to HomepageA Planning Resource Built in Penang, Used Across Malaysia
Wexar was founded in Penang with a focus on one specific gap in the AI planning landscape: the space between a team deciding to adopt AI workloads and the point where infrastructure procurement begins. That middle ground — where assumptions need to be documented, compared, and communicated — had no dedicated tooling.
We started with a simple structured template, refined through conversations with engineering leads and infrastructure teams at organisations of different sizes. The feedback shaped both the content and the format: clear cells, traceable assumptions, and outputs that non-technical stakeholders can review without a walkthrough.
Today, Wexar offers three resources at different scopes — from a single-seat template to a custom-built model. All of them share the same design principle: compute planning should be readable as a grid, not buried in a document.
Our Mission
To give planning teams a clear, shared language for AI compute requirements — so that the assumptions behind every workload are visible, comparable, and easy to revisit.
Our Approach
We design every resource around the planning grid as a visual metaphor. Rows for workload types, columns for parameters, and a clear separation between assumptions and outputs — so teams can debate the inputs rather than rebuild the structure.
Our Commitment
We document our assumptions, keep our language plain, and build resources that remain useful as AI workloads evolve. When a resource is no longer fit for purpose, we update it rather than issue a new one at the same price.
The People Behind Wexar
A small team with a focused scope — planning methodology, technical writing, and client engagement.
Ahmad Zulfikri
Co-Founder & Planning Lead
Former infrastructure planning lead with experience across telecommunications and data centre projects in Peninsular Malaysia. Designs the core planning framework behind each Wexar resource.
Nurul Rashidah
Co-Founder & Documentation Lead
Technical writer specialising in structured documentation for engineering teams. Responsible for the guides, worked examples, and handover materials included in every resource.
Kelvin Wong
Client Engagement
Manages discovery sessions and client communication for the Bespoke Planning Model. Background in project coordination across Penang's technology sector.
How We Maintain Quality Across Every Resource
Each resource goes through a defined review process before delivery. These are the standards we apply internally and communicate to our clients.
Assumption Traceability
Every figure in our resources is tied to a documented assumption. We treat an assumption that cannot be sourced as a gap, not a default.
Data Handling Standards
Client information shared during discovery sessions is held only for the duration of the engagement and not retained or shared beyond delivery. Aligned with Malaysian PDPA principles.
Scheduled Reviews
Standard resources are reviewed against the current AI infrastructure landscape every six months. Clients who have purchased a resource are notified when a significant update is released.
Clear Documentation
Every resource includes a usage guide written for planning professionals rather than only for infrastructure engineers. If a section requires specialist knowledge to use, we document what that knowledge is.
Peer Review Process
Each template and toolkit is reviewed by at least one team member who did not author it, with a focus on internal consistency, usability, and language clarity before release.
Post-Delivery Support
Clients can submit questions about their purchased resource within 30 days of delivery. We respond within two business days during that period.
Why Structure Matters in AI Compute Planning
AI workload planning sits at the intersection of engineering, finance, and operational strategy. The decisions made at this stage — about peak demand assumptions, utilisation targets, and scaling thresholds — shape procurement, staffing, and project timelines for months or years ahead. Yet the tools teams reach for are often general-purpose: spreadsheets with inconsistent structures, documents that do not distinguish inputs from derived figures, or slide presentations that present conclusions without the underlying logic.
Wexar is focused on this specific planning stage. Our resources are not infrastructure management platforms or monitoring dashboards. They are planning aids: structured frameworks for documenting what your team believes about your AI workloads before you commit to hardware or cloud contracts. That distinction matters because the questions being answered are different — not "what is the system doing right now" but "what do we need the system to handle, and under what set of assumptions did we reach that figure."
Penang's position as a hub for electronics manufacturing, semiconductor production, and emerging technology development means there is a particular concentration of organisations working through exactly these questions. Wexar is based here deliberately, drawing on that context to keep our planning frameworks relevant to the kinds of AI deployments taking shape in the region — from industrial process optimisation to logistics routing to enterprise language model deployment.
Our values are reflected in how we build: we document assumptions rather than embedding them silently, we keep language accessible rather than relying on jargon, and we build resources that remain usable after the initial planning session rather than becoming reference documents that nobody returns to. That is the standard we hold ourselves to with every resource we release.
Interested in What Wexar Can Offer Your Team?
Reach out to discuss your planning scope. We will help you identify which resource suits your team's current needs.
Get in Touch