I'm interested in the interface between high-level abstraction and low-level systems. I often write about what many overlook—internal implementations, OS behavior, memory layout, and boundary cases.
I specialize in bridging infrastructure, machine learning, and automation to improve system reliability and operational efficiency across diverse industries.
I have contributed to foundational infrastructure projects in AWS environments, implementing recovery automation via Terraform and improving CI/CD pipelines by eliminating manual inefficiencies. My involvement in ISMS and PMS committees has also supported the successful alignment of operational frameworks with ISO27001 and ISO27017 standards.
In machine learning, I have developed time-series analysis systems for contribution analysis and deployed real-time object detection solutions using CNNs and Transformers. These projects included full-cycle development—from data preparation to model tuning and evaluation—integrated within MLOps pipelines both in cloud and on-premise environments.
My work has extended to robotics and wireless communication, where I built edge computing solutions and AWS-integrated monitoring systems that enhanced field operation efficiency. Additionally, I have taught machine learning internally, fostering cross-functional understanding and capability building.
I place high value on quiet precision over hype, and focus on building reliable, maintainable systems informed by hands-on operational insight and deep technical grounding.
I write about Ruby internals, system calls, struct packing in C, and low-level logic behind high-level languages. Articles are mostly in Japanese.
👉 Technical Blog Archive (English)
👉 View my articles on Zenn (Japanese)
I'm not actively seeking attention, but if you'd like to reach out for collaboration or discussion, please open an issue on my GitHub repository.
📧 Email: kazuki.kanda05570[at]gmail.com
Thanks for reading.