I help teams turn imaging concepts into testable camera prototypes—aligning system architecture, sensor integration, electronics, experiments, and graph-based data analysis around the decisions that matter.
A successful imaging prototype depends on the full signal chain: the sensor, readout electronics, timing and synchronization, optics, calibration, data interfaces, and the algorithms that convert measurements into useful information. I connect these layers so hardware choices remain traceable to reconstruction quality and machine-vision performance.
Translate application requirements into an end-to-end architecture, define interfaces and trade studies, plan integration, and drive a prototype from concept through bring-up and experimental validation.
Work directly with CMOS and SPAD image-sensor designers to align sensor behavior, readout modes, timing, data formats, operating constraints, and prototype requirements. Characterization is focused on the measurements needed to make architecture and integration decisions.
Develop and coordinate schematics, component selection, PCB design, power and signal interfaces, design reviews, bring-up plans, and bench debugging for camera and sensor-support electronics.
Model measurements as signals on graphs to fuse heterogeneous modalities, reconstruct missing or sparse data, improve spatial resolution, suppress noise, and preserve meaningful boundaries using physical and contextual relationships.
Structure milestones, coordinate cross-functional contributors, manage technical risk, document decisions, and keep experiments and hardware development connected to a clear validation plan.
Help design, execute, and document controlled technical experiments that support independent expert analysis. Scope, methods, uncertainty, and results remain explicit and reproducible.
My background spans camera and instrument architecture, image-sensor integration, PCB and electronics development, prototype bring-up, experimental validation, and computational imaging. I work closely with image-sensor developers so the prototype architecture reflects how the sensor actually behaves—not only what a data sheet suggests.
Collaboration continues through interface definition, operating-mode selection, timing and synchronization, data acquisition, focused characterization, troubleshooting, and validation. I translate between sensor design, electrical implementation, system requirements, and downstream reconstruction so teams can make informed tradeoffs early.
My doctoral research at the University of Southern California applies GSP and graph learning to multimodal remote-sensing data. The same core ideas—using relationships, context, and physics to recover information from sparse, noisy, or mismatched measurements—extend naturally to SPAD, CMOS, depth, transient, spectral, and other multimodal camera systems.
I lead technical work by making interfaces, ownership, risks, and validation criteria explicit—while remaining hands-on with design reviews, PCB execution, bring-up, data analysis, and experimental planning. I also support controlled experiments for expert-witness engagements, with emphasis on reproducible methods, traceable assumptions, uncertainty, and clear documentation.
Through the NSF I-Corps National program, I developed practical experience in customer discovery and technology commercialization. That perspective helps me connect technical decisions to user needs, development risk, and a credible path from prototype to product.
Tell me what you are building—and where the system is getting stuck.