About Han

Han Cheol Moon


Staff AI/LLM Engineer with a Ph.D. in Computer Science (NTU-NLP Group), specializing in large language models, robustness, and applied NLP systems. I combine cutting-edge research with production-grade deployment expertise, having led the development of NL2SQL systems, enterprise-scale document QA pipelines, and LLM safety frameworks at Samsung AI Center. Experienced in bridging academic research (ACL, KDD, EMNLP publications) with scalable LLM-based services, I bring deep technical knowledge, strong system design skills, and a proven record of delivering reliable, high-impact AI solutions.

Research & Engineering Interests

  • Natural Language Processing: Large Language Models (LLM), LLM Robustness, NL2SQL
  • Machine/Deep Learning: Reinforcement Learning, Generative Models, Kernel Methods
  • Engineering & Systems: LLM Ops, LLM-based Services, Scalable API Design for ML/DL Services

Professional Employments

  • Samsung Electronics, AI Center β€” Staff ML/DL Engineer (Sept 2023 – Present)
    • Led the design and deployment of NL2SQL systems converting natural language queries into SQL, with agentic workflows, schema linking, and fine-tuning on custom databases.
    • Architected scalable document-based question-answering and enterprise translation/summarization services using large language models, with pipelines for parsing, chunking, vector indexing, and API controls.
    • Designed and implemented sensitive data leakage prevention systems, including prompt filtering, model output analysis, and red-teaming strategies, integrated into production AI pipelines.
    • Managed academic collaborations with Seoul National University, supervising student researchers and organizing joint workshops.

Education

  • Nanyang Technological University, Singapore β€” Ph.D. in Computer Science (2019–2023)
  • Yonsei University, Korea β€” M.S. in Electrical & Electronic Engineering (2016–2018)
  • Chung-Ang University, Korea β€” B.S. in Electrical & Electronic Engineering (2009–2016)
    • Completed mandatory Military Service (2009–2012) during studies

Publications

  • Han Cheol Moon and Shafiq Joty. Reinforced Momentum Update for Textual Adversarial Attack, 2024. (In Progress)
  • Han Cheol Moon. Toward Robust Natural Language Systems, Ph.D. Thesis, 2023.
  • Han Cheol Moon, Shafiq Joty, Ruochen Zhao, Megh Thakkar, and Xu Chi. Randomized Smoothing with Masked Inference for Adversarially Robust Text Classifications. In Proceedings of the Association for Computational Linguistics (ACL ’23), Toronto, Canada, 2023.
  • Han Cheol Moon, Shafiq Joty, and Xu Chi. GradMask: Gradient-Guided Token Masking for Textual Adversarial Example Detection. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’22), Washington, DC, ACM, August 14–18, 2022.
  • Junsik Jung, Han-Cheol Moon, Jooyoung Kim, Donghyun Kim, and Kar-Ann Toh. Wi-Fi Based User Identification Using In-Air Handwritten Signature. In IEEE Access, vol. 9, pp. 53548–53565, 2021.
  • Han Cheol Moon, Tasnim Mohiuddin, Shafiq Joty, and Chi Xu. A Unified Neural Coherence Model. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 2262–2272, Hong Kong, China, 2019.
  • Han-Cheol Moon, Se-In Jang, Kangrok Oh, and Kar-Ann Toh. An In-Air Signature Verification System Using Wi-Fi Signals. In Proceedings of the 2017 4th International Conference on Biomedical and Bioinformatics Engineering (ICBBE 2017), 2017.

Skills

Strong background in ML research and scalable LLM system development, with expertise in PyTorch, FastAPI, LangChain, Redis, RabbitMQ, and vector databases.

  • ML/DL Research: PyTorch, TensorFlow, HuggingFace, SciPy, NumPy, Scikit-learn, Pandas
  • ML/LLM System Development: LangChain / LangGraph, Pydantic AI, FastAPI, Redis, RabbitMQ, K6, Ansible, Vagrant
  • Databases: PostgreSQL, MongoDB, SQLite (SQLAlchemy, Alembic), Vector Databases (Qdrant)
  • Programming: Python (primary), Go, C/C++, MATLAB, Bash/Shell
  • Workflow & Tools: Linux (Debian & Arch), Docker, Git, Async/Multiprocessing, LaTeX, Vim + i3
  • Other: Unit Testing, PlantUML, Documentation, Presentations, GIMP, Inkscape, Hugo
  • Soft Skills: Clear Documentation, Engaging Presentations, Team Collaboration, Time Management, Problem-Solving

Teaching

  • Internal Lecturer β€” Samsung AI Center (2023–Present)

    • Led a study group and delivered in-depth lectures and workshops on Machine Learning, Reinforcement Learning, Large Language Models (LLMs), and LLM service design for Samsung engineers and researchers.
    • Covered theory, applications, and advanced topics such as policy gradients, prompt engineering, and fine-tuning.
      (See lecture materials).
  • Teaching Assistant β€” Nanyang Technological University, Singapore (2019–2020)

    • DeepNLP β€” Assisted in tutorials, grading, and mentoring students
  • Teaching Assistant β€” Yonsei University, Korea

    • Digital Logic Circuits (Spring 2018) β€” Assisted in tutorials and grading
    • Multimedia Signal Processing (Fall 2017) β€” Assisted in labs and coursework support
    • Signals and Systems (Fall 2016) β€” Assisted in tutorials and exam preparation

Honors & Awards

  • Singapore International Graduate Award (SINGA) β€” 2019–2023
  • Top of Class Scholarship, Chung-Ang University β€” 2015
  • Dean’s List, Chung-Ang University β€” 2013–2014

Service & Leadership

  • Republic of Korea Army β€” Capital Defense Command “SHIELD”, 1st Security Group (2009–2012)
    • Served as a Sergeant, guarding the presidential residence and performing security duties
    • Developed discipline, responsibility, and leadership through service

Technical Interests

  • Technical Writing & Knowledge Sharing
    Consistently producing study notes and articles since my master’s program, covering LLM system design, deep learning, mathematics, and programming. I maintain a curated collection on GitHub and regularly share articles on my Blog, demonstrating my long-term commitment to continuous learning and knowledge sharing.

  • Linux Ricing
    Passionate about customizing Linux environments for efficiency and aesthetics. Since 2016, I have explored Arch Linux, i3, and Vim-based setups, creating highly optimized workflows tailored to development and research. This long-term practice reflects both my technical curiosity and my persistence in mastering complex systems.