Resume
Hi, I am

Jinhyuk Jang.

I build Actions from Perception.

I build complete autonomous systems—from low-level robot control to AI perception and central management servers.

Check my work
Jinhyuk Jang

About Me

I am a Full-Stack Robotics Engineer capable of building complete autonomous systems from system design to AI implementation. I bridge the gap between hardware control, AI perception, and software services to create robots that work in the real world, not just in labs.

Education

KAIST
School of Electrical Engineering
2018.03 - 2025.02 (Leave of Absence)
Addinedu IT Academy
ROS2 & AI Autonomous Robot Bootcamp
2025.02 - 2025.08

Tech Stack

ROS 2 (Humble/Jazzy) Python & C++ Nav2 & Manipulation Computer Vision (YOLO/OpenCV) Deep Learning (PyTorch) IoT & Embedded (ESP32) Linux & Docker AWS & MySQL Communication (TCP/REST/Serial)

Work Experience

LK ROBOTICS

Embedded AI Engineer 2025.10 -
  • Developed PyroVision, an early fire detection system using NXP i.MX8M Plus NPU.
  • Implemented RGB-IR Sensor Fusion algorithm to minimize false positives from smoke/fog.
  • Optimized YOLOv8 & TensorFlow Lite models for edge inference on embedded devices.
  • Conducted advanced research on Transformer-based Re-ID (TransReID) and MOT (ByteTrack/OC-SORT) algorithms using YOLOv11.
  • Built real-time monitoring infrastructure using TCP/IP, GStreamer, and PyQt6.

Featured Work

Shopee

Shopee: Robot Shopping System

A multi-robot autonomous shopping system where robots navigate shelves, pick items, and deliver them. Features ROS2 architecture, FSM-based coordination, and central server management.

ROS 2 Nav2 Multi-Robot
Roomie

Roomie: Hotel Delivery Robot

Autonomous hotel service robot with inter-floor navigation using robot arm for elevator control. Features YOLOv8n vision for obstacle and glass door detection. Achieved 95% button recognition accuracy with YOLO+CNN pipeline.

ROS 2 YOLOv8 Robot Arm IoT (ESP32) PyQt
FALCON

FALCON: Airport Safety AI

Real-time bird strike and FOD detection using Blender/Unity synthetic data. Features collision risk calculation and LLM-based pilot voice Q&A.

AI/CV Synthetic Data Unity/Blender LLM YOLOv8
Bikecast

Bikecast: Demand Forecasting

Bike-sharing demand forecasting analyzing 44M trip records with XGBoost. Optimized query performance from hours to tens of seconds using indexing.

Data Analysis XGBoost Pandas
DUST

DUST: Smart Logistics Control

RFID-based IoT platform for integrated control of AGVs, conveyors, and gates. Improved response speed 3x using binary protocol.

IoT FSM TCP/IP
Minecraft

Minecraft with ROS2 & SLAM

A robotics experiment integrating Minecraft with ROS 2. Controls the game character using a physical joystick via micro-ROS and performs real-time SLAM mapping.

ROS 2 micro-ROS SLAM ESP32 Minecraft

Get In Touch

I am always open to discussing new opportunities in Robotics and AI. Whether you have a question or just want to say hi, feel free to reach out!

Say Hello