Inner Portrait is a fun personality discovery app that reveals your core traits through two options: a quick 13-question quiz or AI analysis of your written responses. Based on Jungian psychology (the foundation of MBTI), I've transformed these concepts into eight engaging 'inner portraits' that make personality insights accessible and entertaining.

This is an AI-powered HR chatbot for a financial institution that helps staff quickly access policies and check leave balances through a friendly conversational interface.

For security, I implemented an API proxy to hide backend endpoints and used iframe PostMessage communication to protect sensitive data.

The solution can drastically cut response times for HR queries from days to seconds, reducing workload on the HR team while maintaining the strict confidentiality needed in financial services.

I resolved layout issues and implemented content updates for 10+ templates originally written in Paged.js such as invoices, receipt, shipment and delivery records in the revamp project.

To improve layout control, I established a new template framework using node-canvas that serves as the benchmark for future development.

I also coordinated the architecture build-up and integration with my coworkers to speed up the development progress.

On top of a pre-built chatbot, I revamped the web accessibility to meet the WCAG level 1 standard, from pre-login to post-login UI.

I also implemented ongoing enhancement and campaign-related functionalities from frontend to backend to ensure the chatbot aligned with features on the online banking platform, creating better engagement with end users.

I led and developed two digital signage projects that displayed floor details and IoT data connected via Modbus/BACnet, ensuring buildings from developers met ESG standards.

To prevent layout issues with floor details, I implemented Node-Canvas to generate images that could be reliably looped on the displays.

The solution also included a custom portal for easily updating display content.

I led the development of a mobile visitor management system that revolutionized access control with seamless online/offline functionality on tablets.

As technical lead, I pioneered our first major AWS implementation using Lambda, DynamoDB, and Cognito, creating a cloud-native solution that reduced processing time while enhancing security compliance without hardware dependencies.

I developed a virtual card platform from scratch that managed access card registration from multiple sources, while enabling alternative access methods via BLE or QR codes on mobile devices (Mobile app developed by my colleague).

The system featured a custom RESTful API I built to handle all card-related operations securely, working in coordination with the existing card reader infrastructure.

I extended a Java-based plugin within an Appian BPM workflow that transformed payment files into standardised formats on a daily basis.

By implementing multiple interfaces for a new credit card type, I enabled that company to expand payment options for its customers, increasing transaction flexibility while maintaining the established workflow architecture.