PhD, Caltech
From idea to working prototype; I help companies and researchers transform challenging technical ideas into working prototypes and products. My expertise spans machine learning, electronics, embedded systems, sensing technologies, and system architecture, bridging the gap between concept and implementation.
I was charged with designing and building a robotic imaging system and interface for digitally 'fingerprinting' physical artworks at extremely high resolution. The system involved robotics, computer vision, automated scan planning and acquisition, and data management, including blockchain technologies. I also co-authored several related patents. Pictured to the left is one of the robotic gantries scanning a Picasso at Sotheby's art auction house in New York City, to register its authenticity.
For large-scale flow battery systems used in grid-scale energy storage and critical infrastructure applications, I was contracted to design and build a diagnostics platform for evaluating battery health, performance, and degradation analysis. The system combined custom electronics, embedded control and acquisition, and analysis software to automate testing and characterization, providing detailed insight into battery behavior across a wide range of operating conditions. Pictured to the right are my custom-built circuits designed to sense various battery parameters. These systems are deployed in several National Labs as well as private companies.
I was engaged to design and develop the original computer vision algorithms for a large-field microscopy platform capable of imaging extremely large areas while maintaining microscopic resolution. By combining image stitching, object tracking, and other computational imaging techniques, the system enables the analysis of biological samples at a scale that would be difficult to achieve using traditional microscopy approaches. I also co-authored several papers and patents related to this work.
I am designing and building an electronic chemical sensor using quartz crystal microbalance technology. This project is currently in its early research stage. By combining sensor hardware, signal processing, and machine learning, the platform aims to enable the characterization and classification of odors and other gases that would be difficult to identify using conventional sensing approaches. Pictured to the right is my custom-designed and assembled gas sensing chip, able to detect and transmit multiple sensory channels.
I researched and developed several machine learning techniques for fast and efficient classification and pattern recognition, with a focus on improving accuracy while reducing computational time and energy costs. By combining novel algorithms with practical deployment constraints, the work helped enable high-performance visual recognition systems suitable for resource-constrained environments such as mobile and embedded devices. This was the focus of some of my doctoral thesis and several publications and was selected as a winner of the Qualcomm Innovation Fellowship.
Have a challenging technical idea you'd like to validate or prototype? Whether you need a proof of concept, a solution to a difficult engineering problem, or help exploring a new product or design, let's discuss how I can help.