Feedback
Client Reviews
Nurul Zahirah
Head of Risk, Petaling Jaya
The fraud detection system they built for us identified patterns our existing rules-based approach had been missing entirely. What impressed me most was the analyst handbook — our team was operating the dashboard independently within a week of handover.
18 January 2026Kevin Wong
CTO, Kuala Lumpur
We engaged synaptena for a chatbot to handle our customer enquiries in both English and Bahasa Malaysia. The conversation design phase was really collaborative — they spent proper time understanding our typical customer interactions before building anything. The handoff to our agents works smoothly.
24 January 2026Dr. Priya Menon
Research Director, Cyberjaya
Their data anonymization work was thorough and well-documented. The privacy impact assessment gave our compliance team the confidence they needed to approve the data sharing arrangement. The pipeline they delivered is something we now reuse for new datasets each quarter.
2 February 2026Haziq Azman
Operations Manager, Shah Alam
Honest and straightforward to deal with. When our initial brief was too broad, they helped us narrow the scope to something practical before quoting. The fraud model they delivered has been running for five months now with very few false alarms — fewer than we expected, frankly.
30 January 2026Lee Ting Wei
Digital Lead, Bangsar
Our chatbot now handles about 60% of incoming queries without human intervention. The content update guide they provided means we can add new topics ourselves. Setup took a bit longer than initially estimated, but the end product was worth the wait.
8 February 2026Siti Aisyah
Compliance Officer, Subang Jaya
We needed to share patient data with a research partner without compromising privacy. synaptena's anonymization service gave us a pipeline that our DPO could audit and sign off on. The compliance documentation they provided was genuinely useful — not just boilerplate.
5 February 2026Case Studies
Success Stories
A mid-sized e-commerce platform in KL was losing an estimated RM 180,000 annually to fraudulent transactions. Their existing rule-based system generated too many false positives, overwhelming their review team and delaying legitimate orders.
We built a custom fraud detection model using 18 months of transaction history. The model learned to distinguish genuine purchasing patterns from suspicious ones, with sensitivity configured to the client's tolerance for missed detections versus false alarms.
False positive rate dropped by 72%. The investigation team now reviews 40% fewer cases while catching more genuine fraud. The client estimates recovery of RM 140,000 in the first year of operation.
"The model paid for itself within two months. More importantly, our review team can focus on the cases that actually matter." — Operations Director
A property management company was struggling to respond to tenant enquiries in a timely manner. Their support team was handling repetitive questions about maintenance, payments, and booking of facilities — in both English and Bahasa Malaysia.
synaptena designed a bilingual chatbot trained on two years of support tickets, integrated with WhatsApp and the company's tenant portal. The bot handles standard enquiries and escalates complex issues to human agents with full conversation context.
The chatbot now resolves 58% of enquiries without human involvement. Average response time went from 4 hours to under 2 minutes for common questions. Tenant satisfaction scores improved by 23% over six months.
"Our tenants love getting immediate answers at midnight. And our support team can now focus on the issues that actually need a human touch." — Customer Experience Manager
A healthcare analytics firm needed to share patient datasets with university researchers but could not proceed due to privacy concerns. Manual anonymization was time-consuming and inconsistent, and their compliance team lacked confidence in the process.
We assessed the sensitivity profile of their datasets and implemented a multi-technique anonymization pipeline combining generalization, masking, and synthetic substitution. The pipeline was designed to be reusable for future datasets with minimal configuration.
The anonymized datasets passed external privacy audit review. Data utility was preserved at 94% for the research use case. Processing time for new datasets was reduced from two weeks of manual work to approximately 3 hours using the pipeline.
"For the first time, we have a repeatable process our compliance team trusts. That alone was worth the investment." — Chief Data Officer
Reach Us
Contact Information
59200 Kuala Lumpur
Sat: 10AM – 2PM
Track Record
By the Numbers
Your organisation could be next
Let us discuss how synaptena's AI solutions can address your specific challenges. We would be glad to share more examples relevant to your industry.
Start a Conversation