A research-focused engineer specializing in Computer Vision, Signal Processing, and Deep Learning with a strong academic portfolio. Demonstrates high proficiency in implementing complex mathematical algorithms for Radar and HDR imaging using Python, while also possessing capability in C# and Unity for VR applications. Their work prioritizes rapid prototyping and theoretical depth, though production engineering practices like testing and dependency management are areas for growth.
Classification of Human Movement using mmWave FMCW Radar Micro-Doppler Signature
Model-Agnostic Meta-Learning for HDR Image Reconstruction. By learning the common structure between all LDR-to-HDR conversion tasks, our model is able to adapt it's predictions given extra exposures of a scene. This novel approach reframes LDR-to-HDR conversion as a meta-learning problem.
Object Classification via Range Doppler Plots on PMCW Radar Data
Code is optimized for generating research results and paper artifacts rather than long-term maintenance.
Good separation of concerns in projects like MetaHDR and Blackjack, though code duplication exists in Borealis.
Frequent use of hardcoded absolute paths, debug artifacts (pdb), and lack of dependency files limits portability.
Primary language used across most repositories for complex tasks like meta-learning, DSP, and reinforcement learning.
Demonstrated deep expertise in mmWave radar processing (uDoppler, Borealis) and HDR image reconstruction (MetaHDR).
Implements advanced concepts like Model-Agnostic Meta-Learning and custom loss functions, though code often lacks scalability.
Capable of building functional VR environments (ee267_viper), though relies on legacy assets and hardware-coupled logic.
Unit tests are consistently absent across all major projects, posing high regression risks.
Projects feature excellent high-level READMEs with theoretical context and visualizations, though setup instructions can be brittle.
Get docs, diagrams, scorecards, and reviews for any repository. Understand code faster.