About

System Engineer at MangoBoost, building GPU cluster management software and DPU/IPU infrastructure for AI datacenters. Interested in heterogeneous computing, GPGPU programming, and distributed systems.

Work Experience

• System Engineer, MangoBoost, Seoul, South Korea

  • Nov 2024 – Present · On-site
  • Building systems software and infrastructure for DPU/IPU-based AI datacenter solutions.
  • Jan 2026 – Present: Developing GPU server cluster management software for the Mango GPU Server platform — turnkey AI systems with near-linear scalability using RDMA/RoCEv2.
  • C++, Go, Network Engineering, RDMA, GPU Cluster Management

• Artificial Intelligence Engineer, CREATZ, Seoul, South Korea

  • Oct 2022 – Nov 2024 · On-site
  • Developed and deployed AI models for various vision tasks.
  • Machine Learning, Software Development, Computer Vision

Projects

• Open Programmable Infrastructure (OPI) Project · Apr 2025 – Present
  • Associated with MangoBoost
  • Community-driven, standards-based open ecosystem for next-generation architectures and frameworks based on DPU/IPU-like technologies.
  • Implemented the bridge for MangoBoost products within the OPI project.
  • Go, C++, Network Engineering, OpenStack

Education

• Seoul National University, Seoul, South Korea
  • Master’s degree, Computer Science and Engineering (Sep 2020 – Aug 2022)
    • High Performance Computing (HPC), CUDA
    • THUNDER Research Group · Specialized in Heterogeneous Computing
  • Bachelor of Engineering, Computer Science and Engineering (Mar 2014 – Aug 2020)
    • Machine Learning, Software Development
  • Bachelor of Arts, Philosophy (Mar 2014 – Aug 2020)

Technical Skills

• GPU & Cluster: GPU Cluster Management, RDMA, RoCEv2, CUDA, HIP, SYCL, ROCm, Heterogeneous Computing
• Systems & Infrastructure: DPU/IPU, NVMe/RDMA, Network Engineering, OpenStack, Storage Disaggregation
• AI/ML: LLM Inference, PyTorch, Machine Learning, Computer Vision
• Languages: C/C++, Go, Python, C#