About Me

I'm Wen-Ting (Joseph) Wang, a machine learning engineer with about a decade in production ML and a Ph.D. in Statistics. I build end2end systems, from modeling to data and infra, and make them run in production. Lately I've been working on agentic workflows, RAG, and geospatial/spatiotemporal deep learning. In academic projects, I'm exploring conditional diffusion models and spatiotemporal deep learning methods.

Experience

AI Product Engineering

Built and scaled an entity extraction service from ~2k to ~400k articles per day, powering chat/analytics assistants and a news reporting SaaS; delivered a RAG-based assistant for an energy client in the UAE.

E-commerce

Built large-scale recommendation systems from scratch.

Earlier

Data science/ML roles focused on NLP and time-series modeling.

In Progress

Interpretable low-rank geospatial modeling

Working on improving diffusion models with Bayesian techniques