I like problems that sit between a model and a decision.

I’m a data scientist focused on GenAI applications, rigorous experimentation, and the systems that turn uncertain model output into useful work.

My work has ranged from production RAG and evaluation harnesses to real-time propensity models and advertising experiments. The common thread is translating technical complexity into a decision someone can trust.

I’m interested in what happens after the demo: how a system is evaluated, where authority lives, what failure looks like, and whether the result changes a real business outcome.

AI 对我来说不是信仰，也不是身份标签。它是一个帮助我整理混乱、扩大视野、推进想法的工具。

Connect on LinkedIn ↗