I launched the first in‑house software solution at Samsung Foundry to integrate foundry technology and data science, and I developed a web‑based platform from scratch to support diverse engineering data and enable emerging‑technology skill development.

When I developed the SRAM bit‑cell yield model at Samsung, I found it challenging to help engineers understand bit‑cell operation, statistics, and the long computational workflow involved. To address this, I developed an autonomous simulation tool called Zimulator, utilizing HSPICE, Perl, and Motif on Linux. It helped engineers 1️⃣ quickly learn both the old models and the new model, and 2️⃣ speed up computational pipelines. I faced a similar challenge later when analyzing embedded SRAM yield and logic yield, which led me to realize that I needed an engineering web platform to support this work.

So I took the initiative to launch a new software team with ownership over an integrated data and analysis (IDA) platform, designed to serve as an ecosystem for engineers. The platform was designed to automate control analysis and reporting through statistical analysis, machine learning, and deep learning techniques. I developed over 20 million lines of Python code for the research and development of the web platform. The platform had over 200 subscribers in the Samsung Foundry Technology Development team.

My in‑house software development followed a DevOps model. The process was agile because the developers understood both domains. The platform was built using Node.js, React, HTML, CSS, Bootstrap, jQuery, XML, JavaScript, Python, MongoDB, Docker, and Git.