Some of the projects I’ve worked on in computer vision, machine learning, and autonomous systems.
Current role
I work with engineers and applied scientists on foundation models for autonomous decision-making, from sensors to planning.
Our team is developing Multi-Modal Language Action models (MLAs). My responsibilities include model architecture, training, datasets, benchmarking, and research planning.
Autonomous stores
I worked on Just Walk Out applied science for autonomous stores.
Our team focused on multi-modal models for action recognition.
September 2016 – July 2021
I worked with the R&D team on detect-and-avoid systems for commercial drones. Our camera-based system could detect a small plane with a 10-meter wingspan at about 1.2 kilometers.
This work was recognized by MIT Innovators Under 35 in 2017.
MIT Innovator Under 35
January 2019 – December 2020
I co-founded View-Factor in October 2018 to work on traffic applications of computer vision and machine learning.
We worked on vehicle detection, traffic signal coordination, and traffic analytics.
September 2014 – August 2016
I worked with the R&D team on virtual try-on, including body reconstruction from images, fitting simulation, and rendering.
The project used smartphone photos for body reconstruction. The fitting simulation and rendering were designed to run in less than a second.
