Full Stack Software Developer Intern

• Skills: Next.js, TypeScript, Tailwind CSS, Google Cloud Platform, Microsoft SQL Server, Cloud Run, HTTPS load balancer, CDN, Docker, Prisma
• Migrated a $50M annual-revenue WordPress site to Next.js on GCP, improving Lighthouse performance and SEO scores from the 40s to the 90s through image optimization and an HTTPS load balancer with Cloud CDN
• Launched a new AI chatbot service by integrating internal APIs, adding SMS verification, and redesigning the UI with resilient loading states to handle slow model responses, reducing pre-chat drop-offs, and extending time-on-chat by 500%
• Replaced manual App Engine releases by implementing CI/CD using Cloud Build for automated rollouts to Cloud Run with Cloud CDN, eliminating manual deployment steps and enabling smaller, more frequent, and more reliable releases

      Trainee(Intern)

• Skills: Spring Framework, Spring Boot, Java, Hibernate JPA, MyBatis, MySQL, Vue.js, React, React Native, Django, Flask, Amazon Web Service, Nginx, AWS S3, Docker, Jenkins
• Delivered a travel platform by building a secure Spring, MyBatis, MySQL and Vue.js with Kakao Map, route recommendations, email verification, and personalized features, leading to higher trip-plan creation and content discovery
• Built an AI-powered speech therapy web service with Spring Boot, Flask, and AWS, evaluated a customized open-source baseline and a higher accuracy speech-to-text API to enhance recognition quality
• Launched 100+ Spring RESTful APIs, created 10+ web pages, and managed 3 IP addresses and CI/CD pipelines
• Applied open-source libraries and third-party APIs such as OAuth, OpenAI Whisper, Kakao Map, Korean Government OpenAPIs

      Lieutenant(Signal Officer)

• Led and trained 30-100 soldiers in tactical communications including radios and wired networks, improving coordination and unit readiness during drills and field exercises

      Research Staff(Advisor: Dr. Jeongyeup Paek)

• Led deep learning research on real-time audio-based drone detection and localization, designing server-assisted architecture with embedded sensor nodes streaming audio to a central model
• Constructed a labeled dataset, engineered MFCC features, and evaluated CNN and DNN models, then selected a DNN with multiclass classification for distance estimation, achieving 97% detection accuracy and more reliable ranging
• Demonstrated end-to-end feasibility by processing each one-second audio segment in 0.86 seconds compared with 6.22 seconds on the embedded approach, performed localization via trilateration, and published results at ACM MobiSys 2019

• Developed a system that can distinguish UAV with 97% accuracy using intensive deep learning such based on audio features(MFCCs)

      Cadet Officer(57th, 58th)

      Student Representative

      Teaching Assistant

• Covered basic software programming using Scratch

      Math Lecturer