Ilgyu Yi
Machine Learning & Software Engineer · Seoul, Republic of Korea
Professional Summary
Experienced reinforcement learning engineer with a strong background in simulation-based optimization, combinatorial problem solving, and decentralized system design. Proven ability to define complex technical problems, implement advanced ML architectures, and deliver scalable AI-driven solutions. Passionate about bridging deep tech with impactful business applications.
Work in Progress
Unirec — a modular, RL-friendly recommendation framework with swappable components, typed interfaces, and reproducible caching. github.com/ilgyu-yi/unirec
Professional Experience
Senior Software Engineer — NHN
Republic of Korea · 2025 – Present
- Maintaining and extending Handol.
Blockchain Engineer — Planetarium Labs
South Korea, Singapore · Jun 2022 – May 2025
- Maintained and enhanced Libplanet, an open-source blockchain framework.
- Stabilized consensus mechanisms for PBFT.
- Designed and implemented core consensus mechanisms for DPoS.
- Developed VRF based on RFC 9381.
- Developed probabilistic sortition logic based on the Algorand paper.
ML Engineer — Spacewalk
South Korea · Nov 2019 – Mar 2022
- Led RL-based architectural design automation, from parameterization to optimization.
- Designed autoregressive policies, pointer networks, and generalizable agents across dynamic state-action spaces.
- Launched Landbook Premium mobile app: automatic building design in under 3 min per parcel.
- Collaborated with the product team and architects to continuously evaluate qualitative outcomes.
ML Engineer — Aidentify
South Korea · Sep 2018 – Nov 2019
- Developed an RL agent for HVAC system optimization in data centers.
- Combined MDN, Dyna, and a modified value function for 30% simulated energy savings.
- ML service research projects (2018–2019): domain adaptation (ETRI), neural architecture search (ETRI), anomaly detection (Hyundai Autron), image segmentation (Ministry of Environment).
Supervisor — Hyundai Steel
South Korea · Jul 2017 – Sep 2017
- Full-time Supervisor of the Raw Materials Team.
- Scrap inspection and purchase.
- Establishment of the scrap input plan.
Project Engineer — DL E&C
South Korea · Jan 2014 – Aug 2014
- Project Engineer, Mechanical Engineering Team.
- Scheduling (manpower mobilization plan, engineering plan).
- Cost estimation (mechanical instruments, pipes, manpower).
- Expediting (mechanical instruments).
Education
Yonsei University, BSc in Mechanical Engineering (2008 – 2014)
Technical Skills
Languages: Python, C#
Frameworks: PyTorch, TensorFlow, Pandas, Numpy, Scikit-learn
Domains: Reinforcement Learning, Machine Learning, Probabilistic Modeling, Combinatorial Optimization, Simulation, Distributed Ledger, Open Source
Tools: Git, Docker, Rhino/Grasshopper, EnergyPlus, Unity