Skip to content
View jaehyun429's full-sized avatar
  • ChungNam National University
  • Daejeon

Block or report jaehyun429

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
jaehyun429/README.md

정재현 | Jaehyun Jeong

AI Agent Engineer

Chungnam National University, Computer Science (졸업 예정 2027.02)

Email GitHub


🔬 About Me

LLM Fine-tuning과 Browser Automation을 연구하는 AI Agent Engineer 지망생입니다.

  • 🎓 충남대학교 컴퓨터공학과 4학년 (이영석 교수 Data Network 연구실)
  • 🤖 BUA (Browser Use Agent): QLoRA fine-tuned Qwen2.5-32B로 6개 캠퍼스 시스템 자동화
  • 🎙️ 라디오 방송 구조화 파이프라인: WhisperX + pyannote + Demucs → Interspeech 2026 제출
  • 🏆 KCC 2026 논문 제출 완료
  • 📍 Daejeon, Korea

🛠️ Tech Stack

LLM / AI

Python PyTorch HuggingFace vLLM

Backend / Infrastructure

FastAPI WebSocket LangChain


📌 Featured Projects

🤖 BUA — Browser Use Agent

QLoRA fine-tuned Qwen2.5-32B 기반 캠퍼스 업무 자동화 에이전트

  • CDP(Chrome DevTools Protocol) 기반 브라우저 제어
  • Teacher(Gemini 2.5 Flash) → Student(Qwen2.5-32B) knowledge distillation
  • QLoRA (r=16, α=32), 20개 novel task variants에서 82% task success rate
  • FastAPI + WebSocket 2-pane 실시간 UI

🎙️ Korean Radio Broadcast Auto-Structuring Pipeline

방송 음성 → 구조화 텍스트 자동 변환 파이프라인

  • WhisperX (ASR) + pyannote (화자분리) + Demucs (음원분리) + Panako (핑거프린팅)
  • Interspeech 2026 제출

📄 Publications & Research

  • 🎙️ Interspeech 2026 제출 — Korean Radio Broadcast Auto-Structuring
  • 📋 KCC 2026 제출 — BUA QLoRA Fine-tuning

Pinned Loading

  1. BUA BUA Public

    QLoRA fine-tuned Qwen2.5-32B browser agent for university LMS | KCC 2026

    Python 1

  2. radio-structuring radio-structuring Public

    Hybrid framework for automatic DJ/Music/Ad segmentation and semantic extraction from long-form Korean radio broadcasts | Interspeech 2026

    Python 3