Hello! I'm Minho Park, an ML Engineering Lead who architects and leads the development of scalable, end-to-end LLM Inference Hardware SDKs. I specialize in efficient model serving and operational optimization, currently driving the development of a vLLM-based framework and a Kubernetes-native SDK. I am skilled in managing and automating the full software development lifecycle and have extensive experience in model optimization, MLOps, and deep learning research at an AI semiconductor company.
- ML Engineering Lead at HyperAccel (Dec 2024 ~ Present)
- Data Scientist at PwC Consulting (Jun 2024 ~ Dec 2024)
- ML Engineer at Neubla Korea Corporation (Jun 2022 ~ May 2024)
- ML Engineer at Samsung Electronics (Feb 2021 ~ May 2022)
- Tech Lead at Tmax AI (Feb 2017 ~ Jan 2021)
Leadership & Platform Architecture
- Team Leadership & Management: Leading ML engineering teams to develop and deliver cutting-edge LLM Inference Engine.
- LLM Inference Engine Architecture: Architecting and spearheading the development of LLM Inference Engine (vLLM Hardware PlugIn), and Kubernetes Device PlugIn for HyperAccel LPU.
- DevOps & MLOps Strategy: Managing and automating the entire software development environment to maximize efficiency from development to operation.
- Cross-functional Collaboration: Experienced in working closely with hardware design and development teams to co-create algorithms, design architectures, and patents.
Technical Expertise
- LLM Inference Optimization:
- Distributed & MLOps Infrastructure:
- Languages: Python, C++, CUDA
- ML Packages: vLLM, PyTorch, Transformers, Scikit-Learn, Pandas, SciPy, NumPy
- OS/Cloud: AWS, GCP, Kubernetes, Linux, Docker
- Software Engineering: Git, Github, Github Actions, Jira, Confluence
- M.S. in Nuclear Engineering, Seoul National University (Mar 2015 ~ Feb 2017)
- B.S. in Nuclear Engineering, Seoul National University (Mar 2011 ~ Feb 2015)
Last Updated: Oct 1, 2025