Machine Learning Engineer
📞 (+82) 10-9504-8680 | 📧 [email protected] | 🌐 linkedin.com/in/in-young-cho | 📍 Seoul, South Korea

Machine Learning Engineer with over 3 years of professional experience, specializing in developing 2D/3D generative AIs to expand users' creative possibilities. Expertise in designing scalable algorithms to enhance AI workflows, working closely with research scientists, engineers, and product managers to prioritize solutions that maximize user impact. Skilled in Python, PyTorch, Blender, and experienced with cloud infrastructure like Azure and AWS. Passionate about turning ideas into outstanding user experiences and solving business challenges.

Deep Learning Researcher @ KRAFTON Inc.
📍 Seoul, South Korea | Nov. 2022 - Present

• Developed a scalable algorithm for collecting refined training data from large-scale 3D assets and worked with diffusion models using ControlNet and LoRA, enhancing the quality of multi-modal PBR texture generation.
• Reduced mesh processing time from hours to seconds by designing an algorithm to convert unstructured meshes into watertight forms, reducing complexity from cubic to quadratic.
• Optimized a 3D reconstruction model, generating game assets from a single photo in under 30 seconds on consumer devices. This was recognized as a top AI innovation for the upcoming flagship game, inZOI.

Machine Learning Engineer @ Spacewalk Inc.
📍 Seoul, South Korea | Jul. 2021 - Sep. 2022

• Designed and deployed Point Transformer-based reinforcement learning algorithms (A3C, PPO) for optimizing building designs, improving profit maximization while complying with regulations.
• Applied pairwise ranking learning for reward function tuning, reducing tuning time by 200x and increasing agreement with architect preferences from 55% to 90%, doubling AI outputs meeting human-quality standards to 60%.

MSc in Computer Science @ KAIST
📍 Daejeon, South Korea | Sep. 2019 - Aug. 2021

• GPA: 4.25/4.30
• Thesis: Deep Light Clustering for Denoising Monte Carlo Renderings
• Advisor: Prof. Sung-Eui Yoon
• Relevant coursework: Advanced computer graphics, advanced deep learning, 3D machine learning, GPU programming, design and analysis of algorithms

BSc in Computer Science and Mathematics (Double Major) @ KAIST
📍 Daejeon, South Korea | Mar. 2015 - Aug. 2019

• GPA: 4.00/4.30, summa cum laude
• Honors: Dean's list (top 3%)
• Relevant coursework: Computer graphics, operating systems, real/complex/numerical analysis, linear algebra, matrix computation, probability theory, statistics

1. Rengan Xie, Kai Huang, In-Young Cho, Sen Yang, Wei Chen, Hujun Bao, Wenting Zheng, Rong Li, Yuchi Huo. ReN Human: Learning Relightable Neural Implicit Surfaces for Animatable Human Rendering. ACM Transactions on Graphics, 2024.
2. In-Young Cho, Yuchi Huo, Sung-Eui Yoon. Weakly-Supervised Contrastive Learning in Path Manifold for Monte Carlo Image Reconstruction. ACM Transactions on Graphics (Proceedings of ACM SIGGRAPH), 2021.

1. Sung-Eui Yoon, In-Young Cho, Yuchi Huo. Ray Clustering Learning Method Based on Weakly-Supervised Learning for Denoising Through Ray Tracing. US Patent 12051146B2, 2024.

• Languages: Korean (native), English (professional working proficiency, TOEIC: 910/990)
• Programming Languages: Python, C/C++, CUDA
• Software: Git/Hub/Lab, Linux, Windows, Blender, Jenkins, MkDocs
• Libraries: PyTorch, scikit-learn, NumPy, Pandas, Blender Python API
• Cloud/ML Platforms: Azure, AWS, MLflow, Kubeflow

Last updated: October 2024