Software Engineer @RoboSub
We worked closely with the Raven Studio team to craft a visually stunning and highly functional websites.
Type
Software Sub-team Member
Primary Technologies Used
Python, PyTorch, Yolo, OpenCV, Computer Vision
Owners
RoboSub Club McMaster University
RoboSub: Software Team Member
I am a member of the RoboSub club’s Software subteam, contributing to the development of computer vision systems for autonomous underwater navigation. My work focuses on real-time perception, object detection, and image enhancement to support the submarine’s ability to operate reliably in competition environments.
Team Based Engineering
— Worked within a large student engineering team with defined roles and responsibilities
— Contributed to long-term project development leading up to an annual competition
— Balanced individual ownership with team-wide goals and timelines
Research & Iteration
— Conducted technical research to support software decisions
— Evaluated different approaches before implementation
— Iterated based on testing results and team feedback
Building Software in a Large Team
Developing software in a large, multidisciplinary team introduces challenges such as coordination, version control, communication, and aligning technical decisions across subteams. Managing these challenges effectively is essential for steady progress and reliable system integration.
AI & External Service Integration
Technical Collaboration & Development Workflow
— Training computer vision models for the robot submarine to autonomously navigate
— Trained and fine-tuned YOLOv8 models using PyTorch on COCO-based datasets
— Ran ablation studies on data preprocessing and model configurations to evaluate performance trade-offs
— Integrated model outputs into the broader autonomous system through coordinated interfaces
— Worked within a multi-repository Git/GitHub workflow using branches, pull requests, and code reviews
— Translated research findings into implementation decisions through technical discussions and testing
