Client project results
client feedback

What Teams Say After Working With Us

Engineering leads and product managers on their experience with Skerry Labs — what worked, what surprised them, and what they'd tell other teams.

← Back to Home
reviews

From the Teams We've Worked With

KH

Khairul Hakim

Hardware Lead, Petaling Jaya

"We'd been going back and forth on whether our Raspberry Pi-based sensor node could run the inference locally. The feasibility review answered that in about a week and saved us from a larger investment that wouldn't have worked. Clear write-up, honest assessment."

April 2025 · Edge Feasibility Review

SW

Siew Wei Lim

Engineering Manager, Shah Alam

"The On-Device Setup went smoothly. The update routine they put in place has held up well — our team ran the first model refresh on their own without needing to call Skerry Labs back. That was the goal and it worked. Timeline was roughly what they estimated."

March 2025 · On-Device Setup

RB

Roshan Balakrishnan

CTO, IoT Startup, Cyberjaya

"We needed a fleet of 40 devices running consistent inference. The Deployment Programme was structured in a way that gave our team the review process and the documentation to manage it ourselves going forward. The maintenance handbook in particular has been referenced many times since."

April 2025 · Edge Deployment Programme

NA

Nur Aisyah Zainuddin

Product Lead, Kuala Lumpur

"I appreciated that they told us early on what wouldn't work with our device before spending money on setup. The written summary from the feasibility review was detailed enough that we could take it to our board to justify the next steps. Communication throughout was straightforward."

May 2025 · Edge Feasibility Review

TC

Tan Chin Wei

Senior Engineer, Selangor

"Three weeks for the On-Device Setup, which was in line with what they estimated. The training session was the part we found most valuable — not just being shown what was built, but understanding why the update routine is structured the way it is. That kind of context made a difference."

April 2025 · On-Device Setup

FH

Farah Hanim Mohd Yusof

R&D Director, Klang Valley

"We came in not sure which service was right for us. They helped us figure that out in the initial conversation without pushing for the larger engagement. We started with the feasibility review, which gave us enough clarity to move forward with confidence."

March 2025 · Edge Feasibility Review

case studies

Selected Project Summaries

A closer look at how three engagements developed and what the teams involved took away from them.

Industrial IoT · Selangor Edge Deployment Programme

Fleet of 38 devices, inconsistent cloud dependency

Challenge

A manufacturer running 38 edge sensing devices was routing all analysis through a cloud endpoint due to connectivity gaps at two factory sites. When connectivity dropped, data was queued and processed later — causing 20–40 minute lags in anomaly detection.

Solution

Skerry Labs assessed device specs against the inference workload, configured local builds across all 38 units, and designed a review cycle to monitor drift between device versions. A maintenance handbook covered fleet update procedures step by step.

Outcome

Anomaly detection now runs locally on all devices, independent of connectivity. The engineering team completed their first fleet-wide model update using the documented routine without external support. Engagement ran 7 weeks.

"The review cycle they set up has become part of our standard process now. It's not just something we do because we were told to — it actually catches configuration drift early."

Smart Device · Klang Valley On-Device Setup

Single-device prototype, no clear update path

Challenge

A product team had a working cloud-based inference setup but needed to demonstrate local processing on a prototype for an upcoming investor review. They had the device, but no clear path to get a model running on it reliably.

Solution

The On-Device Setup service took the existing model through a quantisation step to fit device memory constraints, configured the inference runtime, and documented the update procedure so the team could refresh the model ahead of the demo.

Outcome

Prototype demo ran successfully on local inference. Team refreshed the model twice before the review using the documented procedure. Setup completed in two and a half weeks. The team subsequently moved forward with the Deployment Programme for production rollout.

"The training session changed how our team thinks about the device. Before, it was a black box. Now it's a well-understood component with a known maintenance procedure."

by the numbers

Engagement Track Record

4+

Years of edge deployment focus

37

Completed engagements

4.7

Average client rating

94%

Deployments running 12+ months post-handover

Want to Discuss Your Project?

Reach us directly — we're based in Shah Alam and work with teams across Malaysia.

Address

Suite 12-2, UOA Business Park, Shah Alam

Hours

Mon–Fri, 9:00 AM – 6:00 PM

Ready to Start a Conversation?

Tell us about your device, your use case, and where you are in the process. We'll be straightforward about what we can and can't help with.

Get in Touch