← Back to profile

Soojin Jung

Software Engineer

sojjung3@gmail.com GitHub LinkedIn Medium

About

Frontend developer with 3+ years of experience building production-grade web applications in fintech and AI domains. Skilled at architecting scalable component systems and integrating complex backend/AI services into intuitive user interfaces. Recently expanded into backend and AI engineering to deliver end-to-end solutions.

Education

BSc. in Applied Mathematics and Statistics & Economics

Stony Brook, State University of New York — 3.57 GPA

Graduated Dec 2018

Backend & AI Engineering Bootcamp

FastCampus K-Digital Training

Jun 2025 — Dec 2025

Skills

Frontend

TypeScript React Next.js Zustand TanStack Query

Backend & Infra

Spring Boot FastAPI MySQL AWS Docker

AI / ML

LangChain RAG OpenAI API

Work Experience

Frontend Developer

BeHappy

Oct 2024 — May 2025

  • Built an AI-powered mobile finance service with real-time chatbot using SSE and OpenAI Whisper (<500ms latency).
  • Engineered a 4-tier data persistence strategy (Recoil + localStorage) managing 310+ dynamic fields with zero data loss.
  • Led a data-driven product pivot using GA4/Mixpanel, uncovering 95% mobile access rate.

Frontend Developer

aiZENGlobal

Mar 2022 — Jun 2024

  • Architected a tri-portal loan brokerage platform (Customer, Operations, Financial Partners) from the ground up.
  • Built 99+ reusable UI components and a config-based SearchTable pattern, reducing dev time by 60% across 30+ pages.
  • Implemented JWT auto-renewal with Request Queue pattern and 8-stage auth flow for financial data security.

Projects

RAG-based AI Policy Assistant

Aug 2025 — Nov 2025

  • Built an end-to-end RAG pipeline (LangChain + ChromaDB) centralizing 113+ policies; reduced latency 66% via parallel document grading.
  • Engineered a multi-step fallback strategy (RAG → Query Re-write → Web Search) achieving 100% query fulfillment.
LangChain ChromaDB RAG Python

Gacha Shop Map Service

Nov 2025 — Present

  • Architected a 3-layer architecture with Container/Presenter pattern; hybrid state management (TanStack Query + Zustand) across 21 API hooks.
  • Optimized geospatial performance reducing API overhead by 80% and component complexity by 44%.
React TypeScript Zustand TanStack Query

Personal Color Self-Diagnosis Service

Feb 2023 — Aug 2024

  • Scaled to 77,900+ diagnoses and 27,600+ AAU with 91.5% completion rate.
  • Achieved 83% organic traffic (30,000+ sessions) via SEO/SSR optimization, ranking on Google's first page.
React Next.js Recoil SEO

Additional