Vijay Gunasekaran
Machine Learning Engineer @Brambles
Signup · Get unlimited contacts
WORK HISTORY
Machine Learning Engineer @Brambles
Orlando, FL, US
AI Repair Supervision & Safety Monitoring (AIRS) Designed a vision-based system to estimate injury risk for workers and detect the completed pallet-repair activity. Projected to save $80 million in costs. Lead a team of 5 to develop, deploy and monitor the system in production by integrating Object Detection, Pose Estimation, Activity Detection, Ray and PostgreSQL for real-time monitoring of multiple operators on-site. Collaborated with the Safety department to design a customized Ergonomic Risk Assessment method, synthesizing elements from REBA and RAMP II evaluation methods for enhanced safety monitoring of repair operators. Set up a data-driven approach to standardize & optimize the data collection engine based on a multi-pipeline, leveraging iterative active learning cycles to reduce the expected data required by 50%. Executed a comprehensive MVP to assess the feasibility of 3D pose estimation and personnel risk assessment in a factory setting.Protruding Nail - Object Detection Implemented a model for detecting protruding nail defects to improve the quality of pallets and reduce handling injuries. Improved the precision of the model by 5% to 94% using state-of-the-art techniques with YOLOv5. Conducted comprehensive error analysis to identify data issues and model weaknesses, leading to targeted improvements in detection.
EDUCATION
University of Central Florida
Master's degree, Computer Science
Anna University Chennai
Bachelor's degree in Computer Science and Engineering , Computer Science
P.S Senior Secondary School
High School
SKILLS
ABOUT VIJAY GUNASEKARAN
I\'m currently working as a Machine Learning Engineer at Brambles.My work at Brambles involves automating factory processes and improving efficiency at CHEP\'s industrial plants in the US. I\'ve worked on and lead projects on detect defection and safety monitoring respectively. At Oregon State, I worked under Dr. Fuxin Li on Anomaly Detection and Uncertainty in AI.I\'d been a computer vision R&D engineer at Algoface, a subsidary of SilkRoute Global Inc. I\'ve significant coding experience in python and C\\C++. I\'m actively seeking full-time opportunities to build efficient systems.I\'d worked for over a year under Dr. Mubarak Shah on the problem of Image geo-localisation using deep learning. I also worked under Dr. Ulas Bagci on detecting and localising lesions in endoscopy videos. I\'d like to work in a similar role that requires the most of what I can offer and pushes me beyond my comfort zone.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.