Wei-Tai Chen

Summary

With a strong background in computer vision, machine learning, and software management, I am passionate about solving real-world problems using machine vision techniques. With over 10 years of coding experience and 5+ years in software engineering management, I have developed diverse vision systems across academic and industrial applications, from defect detection and 3D vision measurement to robot control and AI vision recognition software.

As a Software Engineering Manager, I specialize in technical communication, Agile project management, secure software development, and software release management. My experience in leading cross-functional teams and delivering high-quality solutions enables me to drive innovation in automation and vision systems, bridging technology and real-world applications.

Skills

Professional Experience

• Led team of 4–6 developers in camera integration and vision recognition software projects.
• Combined AI localization with AOI measurement to measure randomly positioned objects; completed 8 production-line cases last year.
• Redesigned UI with a flowchart-based toolchain allowing direct integration of multiple AI tools, cutting project implementation time by ~40%.

• Implemented GitLab CI/CD automation, reducing release time from ~2 hours to ~12 minutes.
• Integrated unit testing into CI/CD, dropping post-release bugs from ~20/month to ~6/month.
• Adopted Agile/Scrum with a task board; increased average sprint issue resolution from 26 to 45.

• Maintained zero security issues over 5 years via software encryption (Dongle), digital signatures, and Advanced Installer.
• Earned IPAS Certified Information Security Engineer and Google Cybersecurity certifications.
• Led Solomon’s IEC 62443-4-1 certification (Oct 2024), securing the SDLC and strengthening the company’s cybersecurity foundations.

• Upgraded UI from WinForms to WPF, improving visualization and interactivity; +20% user satisfaction (survey).
• Built a multi-layer tools framework (crop + AI tools), expanding accepted use-cases by ~60%.
• Released a cloud-based AI recognition version with API training/detection, raising acceptance rate by ~20%.

• Integrated PLC, TCP/IP, and multiple 2D/3D cameras into software; +50% acceptance rate.
• Added line-scan and thermal cameras to robotic smart picking solutions; demoed at Automation Exhibition Taipei.
• Developed a C# SDK covering annotation, training, and detection to enable SI partners; +50% integration acceptance.

• Vision-Based Quality Control for CNC Thin Film Manufacturing: Created an innovative image thresholding method to detect subtle defects in unimodal histograms (presented at 2018 American Automatic Control Council).
• Quality Measurement for Inkjet-Printed Flexible Sensors: Built an OpenCV + Qt GUI to measure edge sharpness, reflection variance, and connectivity, enabling rapid large-batch evaluations.

• Automated Measurement System for Plant Factories: Invented a hydroponic plant weighing device (US Patent US 9,775,300 B2) and maintained a C++/Qt stereo-vision monitoring GUI; results published in Frontiers in Plant Science (2016).
• Butterfly Recognition App: Built a GrabCut-based background removal and SURF matching pipeline; maintained the Android app in Java (Eclipse).

Education