ML Research Engineer specializing in video understanding, deep learning, and scalable evaluation systems. Experienced in R&D for video anomaly detection, medical image segmentation, and end-to-end ML model evaluation platforms.

• Graduated with honors from Yonsei University (M.S.) and The Catholic University of Korea (B.S.)

• Published as first author at premier international conference, and co-authored numerous publications

• Served as a peer reviewer for renowned international journals, including IEEE Transactions on Image Processing (TIP)

• Curriculum Vitae (CV)

• Developed end-to-end ML model evaluation platforms

  • Benchmarking: Led qualitative and quantitative evaluation and benchmarking of frontier models (e.g., Gemini, GPT), delivering actionable research insights

  • Visualization: Built a web-based research visualization dashboard enabling result exploration, demos, and researcher collaboration

  • Scalable Infrastructure: Architected scalable research infrastructure, including distributed LLM-based QA with Ray, a Feast-based video feature store, and an end-to-end video evaluation pipeline

• Conducted research on video understanding, focusing on video scene segmentation and video RAG

• Company website

• Conducted research on deep learning, medical image segmentation, and video anomaly detection

  • Research portfolio

  • Lecture slides (video anomaly detection)

• Graduation thesis: VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detection

  • Paper

  • Presentation

• Advisor: Prof. Sanghyun Park (sanghyun@yonsei.ac.kr)

  • LAB website

• Conducted research on object detection for smart farm

  • Project

• Advisor: Prof. Hochul Kang (hckang19@catholic.ac.kr)

  • LAB website

• M.S. in Computer Science

• GPA: 4.33 / 4.5

• B.S. in Computer Science

• GPA: 4.17 / 4.5 (Rank: 4/51)

• Peer reviewer for international journals and conferences

  • IEEE Transactions on Image Processing (TIP)

  • Pattern Recognition (PR)

  • AAAI Conference on Artificial Intelligence (AAAI)

• Domains: computer vision, video anomaly detection

• ML - PyTorch, vLLM, Ray, Feast

• Web and Backend - Django, Node.js, Spring, Vercel

• Utilities - LaTeX

• Korean - Native

• English - Conversational and Academic presentation

Accepted at NVIDIA GTC 2026, San Jose, CA

• Enhanced video retrieval performance through scene-based video indexing

• Conducted Video QA experiments comparing uniform segmentation and scene-based segmentation on long-video benchmark

Grand Prize in Capstone Design Contest (Top project)

• Led the Fridge team; designed and implemented an end-to-end AI system

• Developed a vision-based AI chatbot for automatic food expiration-date recognition and management

• Conducted industry interviews with E-mart (Shinsegae Group) to validate real-world adoption scenarios

• AnyAnomaly: Zero-shot Customizable Video Anomaly Detection with LVLM - IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Arizona, USA (2026)

  • Authors: Sunghyun Ahn*, Youngwan Jo*, Kijung Lee, Sein Kwon, Inpyo Hong, and Sanghyun Park

  • Paper · Github · Project

• VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detection - Asian Conference on Computer Vision (ACCV), Hanoi, Vietnam (BK list, IF=1) (2024)

  • Authors: Sunghyun Ahn, Youngwan Jo, Kijung Lee, and Sanghyun Park

  • Paper · Github · Project

• Making Anomalies More Anomalous: Video Anomaly Detection Using a Novel Generator and Destroyer - IEEE Access (SCI(E)) (2024)

  • Authors: Seungkyun Hong*, Sunghyun Ahn*, Youngwan Jo, and Sanghyun Park

  • Paper · Github · Project

• Dual Stream Fusion U-Net Transformers for 3D Medical Image Segmentation - IEEE International Conference on Big Data and Smart Computing (BigComp), Bangkok, Thailand (2024)

  • Authors: Seungkyun Hong*, Sunghyun Ahn*, Youngwan Jo, and Sanghyun Park

  • Paper · Github · Project

• Unified Video Anomaly Detection Model for Detecting Different Anomaly Types - IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Arizona, USA (2026)

  • Authors: Kijung Lee, Youngwan Jo, Sunghyun Ahn, and Sanghyun Park

• AMoE-BTS: An Adaptive Mixture of Experts for Clinical Decision Support in Multimodal Brain Tumor Segmentation - ACM/SIGAPP Symposium On Applied Computing (SAC), Thessaloniki, Greece (BK list, IF=1) (2026)

  • Authors: Jeongeun Kim, Youngwan Jo, Sunghyun Ahn and Sanghyun Park

• GranQ: Efficient Channel-wise Quantization via Vectorized Pre-Scaling for Zero-Shot QAT - ACM/SIGAPP Symposium On Applied Computing (SAC), Thessaloniki, Greece (BK list, IF=1) (2026)

  • Authors: Inpyo Hong, Youngwan Jo, Hyojeong Lee, Sunghyun Ahn, Kijung Lee and Sanghyun Park

  • Paper

• MDVAD: Multimodal Diffusion for Video Anomaly Detection - Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), Sydney, AUS (BK list, IF=1) (2025)

  • Authors: Kijung Lee, Youngwan Jo, Sunghyun Ahn, and Sanghyun Park

  • Paper

• Advanced Knowledge Transfer: Refined Feature Distillation for Zero-Shot Quantization in Edge Computing - ACM/SIGAPP Symposium On Applied Computing (SAC), Sicily, Italy (BK list, IF=1) (2025)

  • Authors: Inpyo Hong, Youngwan Jo, Hyojeong Lee, Sunghyun Ahn, and Sanghyun Park

  • Paper