Skip to content
View CocoRoF's full-sized avatar

Block or report CocoRoF

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 supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
CocoRoF/README.md

Typing SVG





About Me

class AIResearcher:
    def __init__(self):
        self.name = "HaRyeom Jang"
        self.role = "AI Researcher & Full-Stack Developer"
        self.location = "Seoul, South Korea ๐Ÿ‡ฐ๐Ÿ‡ท"
        self.current_focus = [
            "LLM Alignment & RLHF",
            "Efficient Fine-tuning (LoRA/PEFT)",
            "Multi-modal AI Systems"
        ]

    def say_hi(self):
        print("Thanks for dropping by! Let's build something amazing together ๐Ÿš€")

me = AIResearcher()
me.say_hi()

Research Focus & Expertise

๐Ÿง  Core AI Research

Domain Technologies
LLM Alignment PPO, GRPO, RLHF, DPO
Knowledge Transfer KL Divergence, GISTLoss
NLP & Embeddings Sentence Transformers, BERT
Multi-modal AI Vision-Language Models

โšก Applied Technologies

Domain Technologies
Efficient ML LoRA, PEFT, QLoRA
Distributed Training DeepSpeed, FSDP
Graph Analytics GraphSAGE, GNN
Behavioral Modeling Recommendation Systems

Tech Arsenal

๐Ÿ AI/ML & Deep Learning

Python PyTorch HuggingFace TensorFlow DeepSpeed CUDA

๐Ÿ’ป Programming Languages

Python TypeScript C# JavaScript SQL

๐ŸŽจ Frontend & Creative Tools

React Svelte TailwindCSS Unity Stable Diffusion

๐Ÿ“š Publications & Research

๐Ÿ”— Full Portfolio: hrletsgo.me | Models & Datasets: HuggingFace

๐Ÿ“– Published Papers ๐Ÿ—๏ธ Major Projects

2024

  • ๐Ÿ“ˆ Stock Price Prediction via Discussion Forum Analysis (IEEE Access)
  • ๐Ÿ›’ Real-time Live Commerce Consumer Behavior Analysis (Under Review)
  • ๐Ÿจ Controllable GenAI for Hotel Review Management (Under Review)

2023

  • ๐Ÿ’ช Health Recommendation System with GraphSAGE
  • โšก Aggregated Power Resource Modeling

Enterprise Solutions

  • ๐Ÿข Samsung Engineering - Strike Risk Management System
  • โšก Power Demand Forecasting (DLinear + Weather Data)

Research Projects

  • ๐Ÿ“Š RL-based Stock Prediction (PPO/Reward Signals)
  • ๐ŸŽฌ Live Commerce Chat Analysis + Sales Forecasting
  • ๐ŸŽจ Multi-modal GenAI Pipeline Development

๐Ÿ† Awards & Recognition

๐Ÿ—“๏ธ Year ๐ŸŽ–๏ธ Achievement ๐Ÿข Organization
2024 ๐Ÿฅ‡ BI Competition Grand Prize (2nd Consecutive Year) Ministry of Trade, Industry & Energy
2024 ๐Ÿ… Best Paper Award Korea Society of MIS
2023 ๐Ÿฅ‡ Public Data BI Competition Grand Prize Ministry of Trade, Industry & Energy
2023 ๐Ÿฅ‡ ERP Innovation Contest Grand Prize Youngrim Won
2022 ๐ŸŽ“ KHU Fellowship Scholar Kyung Hee University
2018-19 ๐Ÿฅ‡ Literary Contest Grand Prize (3x) Sahmyook University

๐ŸŽค Conference Presentations

๐Ÿ“… Year ๐ŸŽฏ Topic ๐Ÿ›๏ธ Conference
2024 ๐Ÿจ Controllable GenAI System for Hotel Reviews KMIS
2024 ๐Ÿ“ˆ PPO-based Stock Prediction Accuracy Comparison KMIS
2023-24 ๐Ÿ›’ Real-time Consumer Behavior in Live Commerce KIIS
2024 โšก Weather-based Power Demand Forecasting KMA

๐ŸŽฎ Beyond Code

๐ŸŽค Singing ย โ€ขย  ๐ŸŽน Piano ย โ€ขย  ๐ŸŽฎ RPG & Simulation Games ย โ€ขย  โœ๏ธ Prompt Engineering & AI Art

"Creativity fuels innovation, and innovation drives progress."

๐Ÿ“ฌ Let's Connect!



Profile Views

โญ "Building the future, one algorithm at a time" โญ

Pinned Loading

  1. Embedding-Is-All-RAG-Need Embedding-Is-All-RAG-Need Public

    Python

  2. Contextifier Contextifier Public

    Contextify is a document processing library that converts raw documents into AI-understandable context. It analyzes, restructures, and normalizes content so that language models can reason over docโ€ฆ

    Python 3 1

  3. Geny Geny Public

    Autonomous multi-agent system โ€” manage multiple agent sessions, orchestrate autonomous tasks, and visualize it all in a 3D city playground.

    Python 3

  4. f2a f2a Public

    One line of code โ†’ Full statistical analysis + interactive HTML report. 24+ file formats, HuggingFace datasets, 6 languages, 20+ analysis modules, 50+ visualizations.

    HTML 1 1