Skip to content
View kaidenjoonko's full-sized avatar

Block or report kaidenjoonko

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
kaidenjoonko/README.md

Kaiden Joon Ko

Computer Engineering & Computer Science @ USC
Building AI/ML systems for healthcare and mission-critical applications

LinkedInEmail


About

I'm a student researcher and engineer specializing in computer vision, multi-agent AI systems, and biomedical machine learning. My work focuses on translating AI research into practical tools for healthcare diagnostics, emergency response, and clinical applications.

Current focus: Real-time ML inference, agentic AI architectures, biomedical signal processing


Projects

Vita Health

Computer vision system for at-home medical guidance

Real-time guidance for accurate blood pressure measurement using object detection and pose estimation. Trained custom YOLOv8 model achieving 97% mAP on medical device detection, integrated with MediaPipe for pose-based feedback.

Python YOLOv8 MediaPipe PyTorch OpenCV

View Repository →


FireLine

Emergency coordination platform with offline-first architecture

Edge computing platform enabling first responder communication during infrastructure failures. Implements guaranteed message delivery with UUID-based acknowledgment and automatic retry mechanisms.

TypeScript Node.js WebSockets Edge Computing

View Repository →


Experience

Software Engineering Intern • Chevron Corporation
May 2025 - July 2025

  • Architected multi-agent generative AI system using Semantic Kernel, accelerating UI development by 95%
  • Designed cross-agent validation algorithm improving system reliability by 40%
  • Built real-time Angular dashboard for AI component preview and approval

Bioinformatics Research Intern • USC Institute for Technology and Medical Systems
August 2024 - May 2025

  • Developed Python/Tkinter interfaces automating EEG data collection for 100+ patients
  • Engineered feature extraction pipeline improving ML classification accuracy by 10%
  • Reduced manual labeling time by 70% through automated biometric signal processing

Machine Learning Research Intern • UCLA Biomedical AI Lab
May 2024 - August 2024

  • Built OCR model achieving 99% accuracy for retinal cancer biomarker extraction
  • Developed preprocessing pipeline for 400+ medical images using TensorFlow and Keras

Technical Skills

Machine Learning & AI
TensorFlow • PyTorch • Keras • YOLOv8 • MediaPipe • Scikit-learn • OpenCV

Languages
Python • Java • C/C++ • JavaScript • TypeScript • SQL • Go

Web & Cloud
React • Node.js • Angular • Vue.js • Express • PostgreSQL • Docker • Azure • GCP

Tools
Git • Linux • Supabase • RESTful APIs • SpringBoot


Research Interests

  • Computer vision for medical applications
  • Biomedical signal processing and ML
  • Multi-agent AI systems and LLM orchestration
  • Edge ML and real-time inference optimization

Recognition

  • HackSC 2025 Winner
  • CURVE Research Fellowship - USC
  • Asian Pacific Alumni Association Scholar

Open to research collaborations and technical discussions in AI/ML and healthcare tech.

Pinned Loading

  1. Fireline Fireline Public

    A mission-aware, offline-first coordination app that keeps first responders located, connected, and alerted using edge computing when networks degrade

    HTML 1

  2. vita1.0 vita1.0 Public

    Python 2

  3. BrainTumorDetector BrainTumorDetector Public

    Brain Tumor Detection Algorithm with Machine Learning

    Python

  4. owenzengusc/Mental_State_Experiment_Platform owenzengusc/Mental_State_Experiment_Platform Public

    Python 2