About me
I'm a software engineer who builds scalable systems that directly improve product performance and user experience. I've led and contributed to engineering efforts across early-stage startups and high-traffic platforms with 4+ million users, shipping features like matching algorithms, email marketing integrations, and real-time chat systems. Whether I'm redesigning data processing pipelines or architecting recommendation processes, I bring a balance of speed, quality, and impact.
Recently, I've been expanding into AI product development — learning GenAI and building projects that combine real-time systems with intelligent automation. My goal is to engineer the next generation of software that's not only reliable and scalable, but also delivers intelligent, adaptive user experiences.
Work experience
At Pixelic, a YCombinator startup, I developed and maintained core CRM features, including scalable bulk contact import functionality. I led the technical integration of Imweb's site builder platform into Spread, our email marketing service, implementing OAuth flows, automated data synchronization using external APIs, and Stripe billing logic.
Remember is a Series D startup with over 4 million users that aims to be the "LinkedIn of Asia." I worked as a full stack engineer, rebuilding a large-scale asynchronous processing system to reduce runtime by over 90% through batching and parallel execution. I built features for a new customer-facing job board. I also provisioned AWS ECS/EC2 infrastructure using Terraform, improving deployment reliability across the platform.
Watcha is a Series D OTT startup offering services similar to Netflix. At Watcha, I developed W-Algorithm, a recommendation feature that increased content savings by 28% and playtime by 35% within a month of release for Watcha TV users. I also collaborated on building a personalized content feed, contributing directly to user engagement and retention metrics for the streaming platform.
Mozzet services Noondate, a dating app with over 4 million users. I led the development of an ElasticSearch-based recommendation system that recommends a certain number of users matching a combination of keywords. I also refactored a complex user-matching pipeline, cutting processing time by 30% through asynchronous task design. Additionally, I implemented a real-time admin-user chat system to enhance in-app communication and support.
Selected projects
Kiqchestra is an open-source Ruby gem that enhances the power of Sidekiq by introducing a job orchestration framework designed to handle complex workflows with ease and efficiency.