I turn complex data into practical AI products. Today, I’m co-founding a healthcare AI venture in South Korea.
My work sits at the intersection of software engineering, AI, and product thinking.
I studied Software Engineering at Beihang University and spent nearly three years as an AI developer at 42Maru, working across frontend, backend, LLM workflows, and cloud infrastructure.
From building an automated chatbot platform to improving AI quality operations and supporting international B2B onboarding, I’ve learned to connect technical execution with the needs of real users.
A few projects that shaped
how I build.
Contributed to SiteBunny, an automated chatbot builder, across the product interface, backend services, LLM workflows, and administrative tools. Worked with RAG and GPT-based systems, and supported B2B onboarding in Saudi Arabia and Mexico.
Moved LLM QA workflows into a back-office system, combining automated first-pass evaluation with human review. Reduced turnaround from four hours to one hour and worked with Korean, English, and Arabic evaluation sets.
Led a team project focused on OCR performance, reducing processing time from 1.37 seconds to 0.09 seconds. The project received an excellence award.
My graduation thesis explored an intelligent meeting room management system using real-time data. I have also worked on multi-agent pipelines and early product validation.
I’m co-founding a healthcare AI venture in South Korea, bringing together AI engineering, pharmacy domain expertise, and product planning.
I’m building alongside a founder who has received venture funding from Y Combinator, three teammates with pharmacy domain expertise, and a product planner. Together, we’re developing practical solutions for medical and healthcare settings.
We are operating in South Korea and preparing to enter the U.S. and global markets. My focus is on turning domain expertise into useful AI workflows, from data preparation to product development and evaluation.
Explore our focusExploring partnerships that connect data needs with appropriate sources and permitted uses.
Focusing on the quality, structure, and context needed to make medical data useful for AI.
Applying experience in LLM workflows, RAG, and evaluation to healthcare use cases.
React · TypeScript · React Query · Tailwind CSS
FastAPI · Pydantic · MongoDB · Redis · Kafka
GPT-based workflows · RAG · Agents · Docker · GitHub Actions · Azure AKS · Helm