About Me

Hello! Welcome to my blog. I am a Senior Machine Learning Engineer at Lyft, where I work on large-scale, production-grade ML systems focused on rare-event detection, statistical efficiency, and safety analytics. My work involves designing and deploying models that extract weak signals from high-volume, highly imbalanced data — often under tight operational and computational constraints.

I spend much of my time thinking about long-tail events, probability calibration, importance sampling, and how to build systems that remain robust under uncertainty. I work across Python, C++, and SQL production pipelines, and collaborate closely with cross-functional teams to ensure that models are not only accurate, but statistically sound and operationally reliable.

In previous roles, I was a Senior AI Engineer at Renesas Electronics, where I implemented deep neural networks on embedded hardware for automotive and industrial systems. That experience gave me a strong appreciation for hardware-aware model design, efficiency constraints, and the mathematics required to make models both light and fast. Earlier in my career, I was an AI Consultant at the Boston Consulting Group and a researcher at the University of Chicago’s Thirty Million Words Center (under John List and Dana Suskind).

I love mathematics, probability, and algorithms, and I use this blog to share thoughts and ideas on these topics. Because of my background in mathematics, I prefer to teach using small toy examples and proofs that make the core idea transparent — often without relying on code. To me, learning AI is about understanding the algorithm deeply enough to implement it by hand with pen and paper.

In my free time I play a lot of basketball and football (soccer), and you can usually find me in rec leagues across the city.