Email: Koichi@taguchi.co.nz
Results-oriented Software and Machine Vision Engineer with over 6 years of experience specialising in computer vision, machine learning, and robotics for manufacturing automation.
Proven expertise in the end-to-end development of industrial visual inspection systems—from problem analysis and AI algorithm design to software implementation, hardware integration, and deployment on the production floor.
Recognised for bridging the gap between cutting-edge R&D (including Vision-Language Models) and practical commercial applications, resulting in three successful technology transfers to business units and three registered patents.
A collaborative team leader and technical mentor who brings strong cross-cultural adaptability, project management skills, and a proven ability to translate complex technologies into customer-centric value.
Python (Proficient); Bash, Shell Scripting, C++, C#, HTML, Markdown
OpenCV, PyTorch, scikit-learn, OpenVINO, Vision-Language Models (VLMs), Large Language Models (LLMs), Image Processing
Docker, Git/GitHub, Linux/Ubuntu, CI/CD (GitHub Actions), AWS (EC2, S3, SageMaker), ROS2, MLflow, Optuna
End-to-End System Integration, 3D CAD (Fusion 360), 3D Printing, Imaging & Illumination System Design
Cross-functional Collaboration, Technical Mentoring, Project Management, Agile/Scrum, Problem-solving
Omron is a global leader in industrial automation, robotics, sensing, and electronic components.
Included role as Technical Team Lead from Oct 2021 – Mar 2022
Successfully executed 3 major technology transfers to commercial business units:
Hosted 6 international students from Australia and the USA for periods up to 9 months. Facilitated daily living support, cultural exchange, and communicated daily in English, demonstrating high adaptability and cross-cultural empathy.
Managed the technical setup, mass-production, and operational debugging of interactive public tech demos (Brain-wave robot control, VR Metaverses, AI image generation).
Contracted by a major food product manufacturer to structure, clean, and analyse complex tabular data to optimise operational workflows.
Developed an automated CI/CD pipeline using GitHub Actions to summarise academic AI papers and post them directly to GitHub Issues.
Mechanical & System Engineering
Chukyo University, Japan (2019)
Google (2025)
Japan Deep Learning Association (2024)
Japan Image Processing Association (2020)
JP2026002147A, 2026 – Lead Inventor
Designed an AR/Visual AI system to assist operators in identifying and matching physical parts to complex assembly instructions.
WO2025142390A1, 2025 – Co-Inventor
Proposed a hash-based nearest-neighbour classification method for fast and accurate decision-making in large datasets.
JP2022163002A1, 2022 – Co-Inventor
Method for reducing ambient light interference when designing optimised lighting patterns for automated visual inspection.
International Workshop on Advanced Image Technology (2019)
Future Convenience Store Challenge at the World Robot Summit (2018)
Co-authored 6 peer-reviewed technical publications and conference presentations on:
Japanese (Native), English (Intermediate – IELTS Academic: 5.0, TOEIC: 545). Highly accustomed to communicating with native English speakers through extensive homestay hosting.
Full Japanese Driver Licence. Eligible for direct conversion to a New Zealand Full Licence.
Team Sports (10+ years playing baseball/softball, emphasising teamwork and communication), Photography (landscapes and family), and exploring international cooking.
Machine Learning Engineer | Computer Vision Specialist | R&D Innovator | Passionate about AI solutions for real-world problems