About

What I build

I am a Machine Learning Engineer and Data Scientist who builds recommender systems end to end, from raw storage to production ranking. At Nam Viet Media I built the recommendation stack for a news platform and a Spotify-style music streaming platform, including audio models that tag songs directly from their waveforms, contributing to an approximately 40% increase in customer retention.

How I work

I build the layer underneath too. Unstructured GA4 event data landed in Amazon S3, refined through a Medallion (Bronze/Silver/Gold) lakehouse on Apache Spark and Apache Iceberg, and orchestrated with Airflow 3 and Prefect on Kubernetes and Docker over Linux servers. I also built the company’s automation layer: about 10 parallel n8n workflows for real-time data verification, a human-in-the-loop LLM content pipeline, and web scrapers on Playwright, Selenium, and the Chrome DevTools Protocol.

What I’m doing now

My grounding is in machine learning, deep learning, and reinforcement learning, with research experience in machine unlearning for recommender systems. I am pursuing an M.S. in Data Science (Machine Learning) at Stony Brook University.