Skip to content
View figure-2's full-sized avatar

Block or report figure-2

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
figure-2/README.md

figure.2

🌱 Hello, I'm Dohyung 🖐

AI Agent · RAG · Multimodal AI

🇰🇷 한국어
경영학을 전공했으며, 현재 통계·데이터과학을 공부하고 있습니다.
AI Agent Engineer를 목표로 RAG, LLM, 멀티모달 AI, Agent Workflow를 중심으로 학습하고 프로젝트를 개발하고 있습니다.

🇺🇸 English
I majored in Business Administration and am currently studying Statistics and Data Science.
I am working toward becoming an AI Agent Engineer, focusing on RAG, LLMs, multimodal AI, and agentic workflows.


🚀 Projects & Learning

📖 학습 내용 · Learning

  • Inch_by_inch
    • 🇰🇷 TIL 정리 및 학습 내용 기록
    • 🇺🇸 Today I Learned (TIL) notes and learning resources

🌟 대표 프로젝트 · Featured Projects

Financial-Insight-Agent · 근거 기반 기업분석

  • 🇰🇷 금융 리포트 RAG, 공시·법령 근거, 시장·포트폴리오 분석을 출처와 함께 통합하는 근거 기반 기업분석 포트폴리오 프로젝트
  • 🇺🇸 An evidence-based company analysis portfolio project integrating financial report RAG, corporate disclosures, legal sources, and market and portfolio analysis with source citations

PlanWeave· 에이전틱 문서 설계

  • 🇰🇷 기획이음(PlanWeave): 아이디어·리서치·기획·편집을 연결하는 LangGraph 기반 에이전틱 문서 설계 도구
  • 🇺🇸 PlanWeave: A LangGraph-based agentic document design tool connecting ideation, research, planning, and editing

SeSAC-Note · 강의 영상 AI 노트

  • 🇰🇷 강의 영상의 음성·화면·시간 정보를 구간별 근거 맥락으로 구성해 AI 노트와 근거 기반 Q&A를 제공하는 멀티모달 AI 서비스
  • 🇺🇸 A multimodal AI service that organizes lecture audio, on-screen content, and timestamps into segment-level context for AI-generated notes and evidence-based Q&A

LocalLens · 로컬 멀티모달 검색

  • 🇰🇷 로컬 텍스트·이미지·PDF를 자연어로 검색하는 멀티모달 Retrieval 서비스
  • 🇺🇸 A multimodal retrieval service for searching local text, images, and PDFs using natural language

🛠️ 기타 프로젝트 · Other Projects


  • History_Docent

    • 🇰🇷 History Docent — 한국사 교육 및 도슨트를 위한 RAG 기반 AI 서비스
    • 🇺🇸 History Docent — A RAG-based AI docent service for Korean history education
  • Insurance_PF

    • 🇰🇷 개인용 자동차 보험 특화 LLM (Auto Insurance Specialist Agent) / LLM 파인튜닝
    • 🇺🇸 An LLM specialized in personal auto insurance (Auto Insurance Specialist Agent) / LLM fine-tuning
  • MULTI_PJT2_4.5HZ

    • 🇰🇷 '4.5HZ' 글로벌 음악 데이터 분석과 사용자 맞춤 음악 추천 서비스
    • 🇺🇸 '4.5HZ' Global music data analysis and personalized music recommendation service

📚 Tech Stack

분야 기술
AI · LLM Python PyTorch Hugging Face Transformers vLLM
Agent · LLM API LangChain LangGraph OpenAI Gemini
Retrieval · Vector Search FAISS ChromaDB pgvector
Multimodal · Document AI SigLIP OpenCV PyMuPDF EasyOCR
Backend · Application FastAPI Pydantic Django Streamlit
Frontend Next.js TypeScript HTML5
Database PostgreSQL Supabase SQLite MySQL
Cloud · Infrastructure Docker AWS Google Cloud
Development Tools Git GitHub Jupyter

적용 기법 · Methods
RAG · BM25 · Multimodal Retrieval · LLM Fine-tuning · Agentic Workflows


🐣 Social

Email   Blog


Pinned Loading

  1. inch_by_inch inch_by_inch Public

    Inch_by_inch

  2. MULTI_PJT2_4.5HZ MULTI_PJT2_4.5HZ Public

    MULTI_PJT2_4.5HZ

    Python