Education

Master of Science, Computer Engineering

Focus areas: Pattern Recognition, Machine Learning, Embedded Systems

Master’s Thesis: Incremental Machine Learning using Support Vector Machines

Designed an embedded machine learning system with Support Vector Machines for tackling Classification and Regression problems on Big Data with accuracies from 82% to 96%
Implemented various incremental learning methods for batch and online training with approximations to reduce run time (-65%), memory (-25%) & computational complexity (-15%)
Implemented PCA, t-SNE and Parallel Coordinate plots for visualizing high dimensional data and K-Means and Nearest Neighbor clustering methods as a test of classifiability
Accelerated training and inference phases by 500% using reconfigurable hardware

[Python] [C/C++] [Verilog] [Scikit-Learn]

Master’s Project: Autonomous Driving on Reconfigurable Hardware

Designed a 1:10 scale self-driving car that can navigate using Computer Vision, aided by sensor fusion of LIDAR and Sonar for Simultaneous Localization and Mapping (SLAM)
System implemented on reconfigurable hardware to emulate over the air firmware updates and designed strategies for protection against external software and hardware attacks

[Python] [C/C++] [Verilog] [OpenCV]

Focus areas: Embedded Systems, Sensor Technology

Bachelor’s Thesis: Intelligent Embedded Myoelectric Controller for Prosthetic Fingers

Inter disciplinary project in association with Texas Instruments and IEEE
Designed a combined Classification and Regression based model for finger movement prediction

[C/C++] [Python] [CAD] [Matlab]