class AIEngineer:
def __init__(self):
self.name = "Hyunbin Kim"
self.role = "AI Backend Engineer"
self.focus = ["LLM Fine-tuning", "Model Deployment", "AI Systems"]
self.background = ["Creative Writing", "Linguistics", "System Design"]
def mission(self):
return "I turn language models into production systems"|
|
|
🎯 MACtuner - LLM Fine-tuning Made Easy
The easiest way to fine-tune LLMs on Mac with Apple Silicon
Core Features
- One-click LLM fine-tuning with LoRA/QLoRA
- Mac M-series optimized (MPS acceleration)
- Visual workflow editor with React + Electron
- HuggingFace integration & GGUF export
- RAG pipeline with PDF support
Tech Stack
| Layer | Technologies |
|---|---|
| Frontend | React · TypeScript · Electron · TailwindCSS |
| Backend | FastAPI · PyTorch · Transformers · PEFT |
| AI | LLaMA · llama.cpp · Sentence Transformers |
🤖 AI Agent System - Multi-agent Orchestration
Autonomous multi-agent system powered by LLMs
Architecture
User Query → Orchestrator → [Agent₁, Agent₂, ... Agentₙ] → Response
↑________________________________↓
Tech Stack: LLM FastAPI RAG Vector Database
🎮 Pressure Project - Real-time AI Application
Real-time AI-powered application with GPU inference
Pipeline
Client ←→ Spring Boot ←→ GPU Inference Engine
(WebSocket)
Tech Stack: Spring Boot WebSocket Docker GPU
📊 Sub Account Stats (bunhine0452)
