David Szczecina

Third year Mechatronics Engineering student at the University of Waterloo.

David Szczecina

Key Areas of Interest

Computer Vision

Bringing pixels to life with intelligent perception

Deep Learning

Unveiling patterns in complexity through multi-layered neural networks

Automation

Empowering efficiency through seamless integration of technology and processes

.

Professional Experience

Backend Engineer,  Magna International, (April - August, 2024)

  • Developed a scalable TCP data processing framework and SQL database for machine state tracking
  • Optimized server for high-volume data using asynchronous connections to handle concurrent clients
  • Integrated AI processing using cloud computing for predictive analytics and downtime prevention
  • Automated defect detection processes using Machine Learning vision systems to reduce headcount

Machine Learning Research Assistant, Vision and Image Processing Group, (November 2023 - Current)

  • Conducting research to better understand global biodiversity through the use of Image Processing
  • Increased Deep Learning model accuracy using Confident Learning for error detection in datasets
  • Performing analysis of how accuracy and loss are affected by mislabeling of common ML datasets

Software Engineering, Toyota Motor Manufacturing, (September - December, 2023)

  • Programed machine vision system to locate features and provide coordinates for 6-axis robot arm
  • Designed vision systems for process automation by utilizing Deep Learning for defect detection
  • Automated audio confirmation processes using Digital Signal Processing to reduce headcount
  • Reduced equipment downtime by performing root cause analysis and reprogramming machinery

Electrical Engineering,  Itipack Systems, (January - April, 2023)

  • Designed, programmed, and tested new electrical controls systems for production line machinery
  • Created high level electrical designs by determining most effective approaches for new projects
  • Iterated on electrical and mechanical designs to increase throughput and improve precision

Product Manager, Zenduit, (September - December, 2023)

  • Developed and implemented AI object detection with YOLO and OpenCV for feature detection
  • Integrated devices using TCP/UDP and MQTT protocols in order to collect location and sensor data
  • Increased accuracy and efficiency by implementing Kalman filters for improved tracking