🎓 ML/DL/CV Engineer/ Researcher | Software Engineer
🔗 Portfolio/Page • LinkedIn • GitHub

Hello! I am a ML/CV Engineer. I did my MSc. in AI at Korea University (2023-25), advised by Dr. Seong-Whan Lee.

I have 4+ years of experience in computer vision, deep learning, robotics and software engineering. My most recent work has been deploying image & video segmentation models.
Aside from Semantic segmentation, I also work on Depth estimation, Instance retrival, Dense matching and Sparse matching. I also have research experience in Big Data and Reinforcement Learning.

• AI/ML: TensorFlow, PyTorch, OpenCV, CUDA, ONNX, Diffusion Models, GANs, Vision Transformers
• Frontend: React, TypeScript, Next.js, HTML/CSS
• Backend: Python, Node.js, C#, C++, MySQL, Bash
• DevOps: GitHub Actions, Docker, NVIDIA Jetson, ONNX
• Domains: Machine Learning, Deep Learning, Computer Vision, Robotics, Edge AI, Image-generation models
• Soft Skills: Problem Solving, Leadership, Team Communication, Academic Writing, Stakeholder Engagement

Ebenezer Tarubinga, Jenifer Kalafatovich, Seong-Whan Lee
A novel approach that addresses coupling, confirmation bias, and boundary blur in semantic segmentation.

Ebenezer Tarubinga, Jenifer Kalafatovich, Seong-Whan Lee
Introduces fuzzy labels and dynamic weighting to handle uncertainty and class imbalance in segmentation tasks.
*Under review

Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning (Liang et al., CVPR 2024).

Semantic-Aware Multi-Label Adversarial Attacks.

Self-Training for Semi-Supervised Semantic Segmentation.

Scalable Urban Dynamic Scenes (Turki et al., CVPR 2023).

Speech Emotion Recognition: self-attention framework for classifying emotions from audio/text).

Autoregressive text-to-image generation.

Controllable text-to-image generation.

Diffusion based text-to-image generation.

Aligning Segment Anything Model to Open Context via Reinforcement Learning.

Tool-Augmented Reward Modeling (Themis) (Li et al., ICLR 2024).

ASD Classification with Multi-Site fMRI Data (Kunda et al., IEEE TMI 2022).

MCF: Mutual Correction Framework for Semi-Supervised Medical Image Segmentation (Wang et al., CVPR 2023).

Why Does the Effective Context Length of LLMs Fall Short? (Chenxin, ICLR 2025).

AI Research Engineer
Gractor (Seoul, Korea)
Sept 2025 – present

• Worked on CV, ML and DL for smart city solutions
• Developed and maintained real-time AIoT systems

AI Research Engineer
PR & ML Lab, Korea University (Seoul, Korea)
Aug 2023 – Aug 2025

• Published research in top-tier venues (IJCNN, Neural Networks).
• Worked on Image and video understanding
• Delivered real-time AI systems and contributed to autonomous driving patent.

Software Engineer
GliT
Jan 2019 – Jan 2021

• Worked with a team of engineers and delivered 10+ successful projects.
• Drove a 35% STEM engagement increase through innovation hubs.
• Developed AI-powered applications using Python, C++, and C#.

• Global Korea Scholarship – South Korea Govt.
• Applied AI Certificate – IBM
• AWS Basics – Amazon
• Modern Robotics Specialization – Northwestern University
• Project Management - Google
• Semantic Segmentation with Amazon Sagemaker - Amazon
• Game Development using Scratch - Massachusetts Institute of Technology
• Machine Learning Pipelines with Azure ML Studio - Microsoft

MSc in Artificial Intelligence
Korea University ( Sept 2023 – Aug 2025

• GINCON Global Committee Member – South Korea National Assembly
• Visit Seoul Foreign Ambassadors – Seoul Metropolitan Government

I'm always open to interesting research collaborations, open-source contributions, or exciting tech initiatives.
Feel free to reach out at ebstarmusic@gmail.com