What Clients Say
About Working with Us
Honest accounts from teams we've worked alongside. These reflect the kinds of engagements we run and what clients found most useful.
Back to HomepageFrom the Teams We've Worked With
The supply chain engagement gave us something we'd been trying to build internally for over a year. The dashboard is actually used — that sounds like a low bar, but many internal BI projects don't clear it. Their documentation made it easy to hand over to our IT team once the engagement wrapped up.
March 2025My main concern was whether the AI outputs would sound like us. Wei Teng spent a lot of time with our editorial team before configuring anything. The first drafts weren't perfect — I wouldn't expect them to be — but close enough for our editors to work with. Setup took slightly longer than expected in week one, but the final result was solid.
February 2025We needed to know whether our credit scoring model was performing fairly across applicants from different states before we scaled. The bias audit gave us a clear picture — some findings expected, a few weren't. The recommendations report was specific enough to act on rather than just generic best practices.
March 2025What I appreciated most was that Razif was honest early on about what was and wasn't possible with the data we had. That directness saves a lot of wasted effort. The predictive model flagged a supplier risk pattern we'd missed for months — the ROI case made itself within the first few weeks of use.
January 2025We'd tried two other AI content tools before this and both produced material that felt generic. The myveltrio setup is different because it's configured specifically to how we write. My writers use it daily now as a starting point, which is exactly what we wanted.
February 2025We'd been using an AI shortlisting tool for eight months before commissioning the audit. The team found patterns we weren't aware of in the training data. The mitigation report gave us a practical path forward — a difficult read in parts, but an important one.
March 2025Three Engagements in Detail
A closer look at how myveltrio's services played out in practice — challenges, approach, and outcomes.
Reducing Unplanned Downtime from Supplier Delays
A mid-sized manufacturer experienced recurring production stoppages from component delivery delays. Procurement had no early warning mechanism — issues only became visible when materials failed to arrive.
myveltrio integrated logistics platform data with supplier communication history and regional port activity. Predictive models flagged at-risk shipments 10–14 days in advance, surfaced through a daily procurement dashboard.
Within the first two months, three potential supply disruptions were identified and mitigated ahead of time. The procurement lead reported fundamentally changed supplier conversations — more proactive, less reactive. Timeline: 10 weeks.
"The dashboard is actually used. That sounds like a low bar, but a lot of internal BI projects don't clear it."
— Head of Procurement, Shah Alam
Scaling Output Without Compromising Editorial Standards
A digital media team needed to increase article output with a fixed headcount. Previous AI writing tools produced content that felt off-brand and required nearly as much editing as writing from scratch.
myveltrio analysed 60 sample articles to extract the publication's voice, then configured a generative pipeline with tuned prompts and guardrails for factual claim verification aligned to their specific editorial style.
Significant reduction in time-per-article in the first month. Editors described AI output as a useful first draft rather than something needing reconstruction. Writers adopted the pipeline as a daily starting point. Timeline: 6 weeks.
"I was sceptical after our previous AI tool experience. This was different — the setup was much more thorough, and the results reflect that."
— Content Director, Petaling Jaya
Auditing an AI Shortlisting Tool Before Wider Deployment
A technology company was preparing to deploy an AI-assisted hiring shortlisting tool across all business units. The People Operations team wanted independent assessment of the model's behaviour before full rollout.
myveltrio created test datasets to isolate the effect of demographic characteristics on model output. Fairness metrics were evaluated across shortlisting rates by gender, age group, and state of origin — covering both model and upstream data.
Significant shortlisting rate disparities found across two of five demographic dimensions. Report provided specific mitigation options ranked by impact. Deployment was paused pending mitigation work. Timeline: 4 weeks.
"Not an easy report to read, but exactly what we needed. We're glad we commissioned it before rolling out further."
— Head of People Operations, Cyberjaya
Ready to Start Your Own Engagement?
We're open to an initial conversation with no obligation. Just tell us a bit about your situation and we'll be straightforward about whether we can help.
Send Us a Message