nuthanan

n
u
t
h

a

4
9
1
n
a
n
t
t
h
h
a
r
m
a
r
a
j
a
h


Software & AI Enthusiast

About Me

I'm a Mechatronics Engineering student at the University of Waterloo with a passion for software development. I enjoy combining my interests in engineering with global affairs and sustainable development.

If you don't see me working or studying, I'm probably at the gym, playing volleyball, biking or travelling.

Feel free to reach out to me through email: n2tharma@uwaterloo.ca

Firmware Developer @ UWFE

September 2024 - Present

Developed firmware for EV control systems and optimized race car dynamics. Collaborated on HIL testing to integrate subsystems and enhance software robustness.

YRHacks Outreach Executive

June 2022 - April 2024

Worked with other YRDSB students to organize a 24-hour hackathon for 300+ students by collaborating with sponsors and community partners.

CANMUN Chief of Staff

December 2022 - June 2024

Led 70+ staff and coordinated a 500+ international level student conference, strengthening project management and team leadership, while ensuring a successful, well-organized event.

Skills

Front-End Development

Worked with HTML, CSS and JavaScript, including packages such as React and next.js.

Machine Learning

Used TensorFlow and OpenCV. Built neural networks for classification tasks.

Hardware & Robotics

Worked with Arduino to build numerous projects applicable in the real world.

Projects

AI Skin Cancer Detection

PyTorch

Programmed a convolutional neural network (CNN) model with the Skin Cancer MNIST: HAM10000 dataset using PyTorch to effectively identify different skin lesion types. Fine-tuned ResNet50 and DenseNet121 models, integrating the custom CNN, and subsequently comparing the performance results between the two models. Achieved a 74% accuracy using the ResNet50 model.

Personal Portfolio

HTML, Tailwind.css, Javascript, GSAP, Anime.js

Developed a platform to highlight my interests and showcase projects from the past four years. Applied UI/UX principles to design a functional and engaging website incorporating animations and cursor-related effects to elevate user experience.

Community Creator

React, CSS, Cohere API

Developed an interactive platform that provides users with AI-powered feedback to improve their community's sustainable practices. Users answer multiple choice questions and receive actionable suggestions to help create a more sustainable environment.

YRHacks 2022 Sustainable Communities Award

Mathematical Analysis of Loss Functions

PyTorch

Conducted research on the mathematical impact of loss functions on the accuracy and efficiency in neural network regression tasks. Concluded that while more complex loss functions improve accuracy, efficiency is reduced.

Mathematical Analysis of Sustainable Architecture

Blender

Designed and modeled a sustainable community using Blender, and analyzed the role that mathematical principles play in architecture and sustainable development.

@2024 Nuthanan Tharmarajah