Reviews
Honest Feedback From People We've Worked With
Piyawan Siriwong
COO, Bangkok · Feb 2026
We brought Umbra Logic in for a readiness assessment and were pleasantly surprised by how thorough and practical their report was. They didn't try to sell us on a bigger project — they just gave us clear, usable findings. Our IT team referenced their document for months afterward when planning our data strategy.
Kittisak Thongsuk
Head of Analytics, Chon Buri · Jan 2026
The custom model they built for our demand forecasting has been running in production for four months now with solid results. The documentation was detailed enough that our internal team was able to retrain it when we added new product categories. My one note: the initial scoping took a bit longer than I expected, but in hindsight it saved us from going in the wrong direction.
Arisa Rattanaporn
Managing Director, Nonthaburi · Feb 2026
What I appreciated most was how they handled our leadership presentation. They translated the technical findings into language our board could actually engage with. No buzzwords, no exaggeration — just an honest picture of what AI could do for us and what we needed to prepare. We've since moved forward with their process mapping engagement.
Somchai Wattanachai
VP of Operations, Samut Prakan · Jan 2026
The process mapping project identified three areas where we were spending significant manual effort on tasks that could be partially automated. Their roadmap was realistic — they told us which ones to tackle first and which ones to hold off on until our data was cleaner. We've already implemented the first phase and are seeing time savings.
Natthakarn Lertpanich
CTO, Bangkok · Feb 2026
Solid work. They were straightforward about the limitations of what our data could support, which I respect. The model they built performs well within the parameters they set, and their monitoring guidelines have been helpful. Communication was responsive throughout the project.
Maneerat Phanich
Director of Strategy, Bangkok · Jan 2026
I'd been approached by several AI consultancies before, but Umbra Logic was the first that actually listened to what our company needed rather than pushing their standard package. They took the time to understand our business before suggesting anything. The readiness assessment report became our internal reference document for the next two quarters of technology planning.
Case Studies
Project Outcomes
Manufacturing Sector — Process Mapping
The Challenge
A mid-sized auto parts manufacturer in Samut Prakan was experiencing bottlenecks in quality inspection, with manual checks consuming significant labor hours. They weren't sure if AI was the right solution or where to begin.
Our Approach
We conducted a five-week process mapping engagement, observing the full production workflow and interviewing floor supervisors and quality leads. We identified three inspection stages where computer vision could assist human inspectors.
The Outcome
The client implemented the first-phase recommendation within three months, reducing manual inspection time by roughly 30% at the identified stages. The phased roadmap continues to guide their technology investment decisions.
Financial Services — Custom Model Development
The Challenge
A regional lending institution needed a more nuanced approach to credit risk evaluation for their SME portfolio. Their existing scoring model was outdated and didn't account for the varied financial profiles of small Thai businesses.
Our Approach
We worked with their data team over eight weeks to build an interpretable gradient-boosted model that incorporated additional features from their transaction data. Every variable was documented with clear business explanations.
The Outcome
Early results after four months in production showed a 15% improvement in prediction accuracy compared to the previous model. The client's risk team was able to retrain the model independently following our documentation.
Healthcare — AI Readiness Assessment
The Challenge
A private hospital group in Bangkok wanted to explore using AI for patient scheduling optimization but had concerns about data quality, staff readiness, and PDPA compliance.
Our Approach
We ran a two-week readiness assessment including interviews with administrative staff, a data quality review of their scheduling systems, and a compliance gap analysis. We presented findings directly to the hospital's executive committee.
The Outcome
The assessment revealed that while the scheduling data was suitable, the organization needed to address data integration between departments first. Our roadmap guided them through a six-month preparation phase before starting model development.
By the Numbers
Our Track Record
6+
Years in Practice
40+
Projects Completed
4.6
Average Rating
68%
Repeat Engagements
Get in Touch
We'd welcome the chance to discuss how we might help your organization.
Considering AI for your organization?
We're here to listen first and advise second. Reach out for a conversation — no strings attached.
Send Us a Message