Jiwon Baek
Full-Stack DeveloperWith 8 SW competition awards and startup frontend intern experience, I've built products spanning RAG·LLM pipelines and AWS infrastructure.
I'm currently an assistant instructor for Samsung executives' AI Agent training. Building primarily on the Claude Code CLI environment, I leverage the latest open-source AI tools to grow as an AI-Driven Developer.
Tech Stack
Work Experience
- Assisted with hands-on AI training for C-suite executives across Samsung affiliates
- Guided practical sessions on workflow automation and website building using the latest LLM tools including Claude Code, Gemini, NotebookLM, and Google AI Studio
- Developed the frontend for a B2B SaaS dashboard
- Refactored repetitive UI into shared components and optimized rendering performance
- Designed and built the UI/UX for a personalized webtoon recommendation service
Award-Winning Projects
GenAI & OCR-based contract risk clause analysis service. Selected as Kyung Hee University's representative team for the national competition
- Designed an OCR and analysis pipeline triggered on contract upload
- Built a masking feature that lets users redact sensitive personal information before upload to prevent data leakage
- Added an editable OCR review step so users can verify and correct recognized text before it feeds the analysis
- Built a Bedrock-based RAG pipeline: S3 storage → Titan embeddings → OpenSearch vector search → Claude 3.5 Sonnet analysis
- Developed the Next.js web interface and iOS WebView integration
- DDOBAK AI analysis scored 14/15 for accuracy via an LLM-as-a-Judge evaluation (vs. 11.25 avg. for commercial LLMs)
Personalized algorithm problem recommendation service based on Baekjoon Online Judge data, built as a React Chrome extension
- Implemented three separate recommendation modes — by problem similarity, forgetting curve, and weak-area model
- Built a Chrome extension so users can get recommended problems directly on Baekjoon
- Won 1st place across all categories at the SW Festival
Awards
Extracurricular Activities
- Completed hands-on labs in Google Cloud study sessions to build cloud infrastructure fundamentals
- Served on the organizing team and led Three.js and React study groups
- Started in CV Track (Cohort 3), later served as NLP Track member and organizer (Cohorts 7–8)
- Studied deep learning and AI fundamentals through Computer Vision and NLP study groups
- Planned and operated data hackathons, handling participant coordination, scheduling, and event logistics
- Completed 6 data/AI projects applying artificial intelligence to real-world problems