Summary
AI engineer and product builder with 8+ years of experience delivering production AI systems across game technology, internal platforms, multimodal QA, retrieval systems, and agentic AI workflows. Currently building end-to-end AI products including multimodal RAG, search-driven agents, and operational AI tooling, while pursuing a Ph.D. in Game Engineering focused on multimodal QA and agent workflows. Strong in translating technical ambiguity into deployable systems, aligning engineering decisions with product outcomes, and working across research, software, and business-facing execution.
Experience
Supercent
Dec 2025 - Present
AI Product Developer
- Designed and delivered production-facing AI workflows, including multimodal RAG systems, search-driven agents, and internal AI tooling.
- Operationalized agentic workflows with emphasis on inspectability, iteration speed, maintainability, and real team adoption.
- Built SAGA (Firebase App Hosting + Genkit), a game-studio knowledge brain — a RAG search engine over game design docs (RAG-Fusion, CRAG, hybrid search, Cross-Encoder reranking, Grounding-Guardian generation; 594 docs / 1,563 vectors across 15 games), in a 3-week build (Feb 2026).
- Built Millie (Firebase App Hosting), a per-user RAG assistant plus a Slack-bot-driven, local-LLM "ClawTeam" autonomous-agent system with an admin control hub for knowledge sources, credentials, agent assignment, and LLM call-quality monitoring, in a 4-week build (late Mar 2026).
SK intellix
Jul 2025 - Dec 2025
AI Manager
- Established AI compliance and technical review structures for smart-home AI capabilities and ecosystem integrations.
- Built operational web tooling and supported roadmap decisions across LLM, computer vision, and monitoring interfaces.
Hongik University
Sep 2024 - Present
Ph.D. Candidate, Game Engineering
- Researching multimodal QA, automated bug reporting, agent workflows, and deployable AI tooling for real production teams.
- Built prototypes with VLMs, retrieval systems, and orchestration frameworks to explore evaluation, tool use, and usable delivery patterns.
NCSOFT
Mar 2018 - Jul 2024
AI Engineer / Technical Builder
- Built AI-enabled systems across gameplay evaluation, procedural generation, automation tooling, and digital human production pipelines.
- Developed simulation-based evaluation agents and internal testing tools to reduce workflow cost and improve team effectiveness.
COM2US
Oct 2016 - Feb 2018
AI / Full-Stack Developer
- Developed reinforcement learning prototypes in game environments and built internal collaboration tooling across frontend, backend, and database layers.
Core Strengths
- Solution architecture and end-to-end AI product delivery
- Agentic AI workflows, tool orchestration, and LLM application design
- Multimodal QA, computer vision workflows, and evaluation-oriented systems
- Retrieval systems, vector search, and AI-assisted interfaces
- Production-minded AI engineering across product, research, and operations
Technologies
- Python, SQL, JavaScript, C#, C/C++, Dart
- PyTorch, TensorFlow, Transformers, OpenCV
- LangGraph, LlamaIndex, RAG, Qdrant, MCP, PEFT/LoRA, ONNX
- AWS, GCP, Docker, Linux, Git, WebSocket, PyQt, Flutter
Education
Hongik University
Ph.D. Candidate in Game Engineering, Sep 2024 - Present
Hongik University
M.S. in Game Engineering, Aug 2014 - Aug 2016
Hongik University
B.S. in Game Engineering, Mar 2007 - Feb 2014
Selected Signals
Patents and Research
2 patents, PeerJ SCI-E paper, HCI Korea paper, IEEE RAAI 2024 poster, 2025 publication on automated game QA reporting.
Leadership and Communication
NCFellowship program management, internal presentations, technology transfer support, and ongoing technical writing on AI workflows.
Open-Source Projects
jeo-code (Bun-based coding-agent harness/CLI framework), jeo-skills (cross-platform agent-skills catalog for harness engineering), and jeo-claw (Discord-controlled, dual-runtime agentic repo-update agent).
Profile
An AI solutions builder with deep game AI production experience and growing strength in agentic, multimodal, and delivery-oriented AI systems.