Turbulence control for drag reduction through deep reinforcement learning
ABOUT & CV
Backend Engineer · Data & AI Researcher
Backend engineer with 4 years building platform backends & infrastructure, plus a 2-year M.S. researching flow control with deep reinforcement learning. I read technology from system fundamentals — network, OS and application layers — and apply it to stable, well-structured backend platforms.
Yonsei University · Environmental & Mechanical engineering
M.S. in Mechanical Engineering.
Thesis — Turbulence control through deep reinforcement learning.
B.S. in Mechanical Engineering. Bachelor in Environmental Engineering (transition from completion to graduation).
Completion of Bachelor in Environmental Engineering.
Korean Augmentation To the United States Army.
Backend & platform engineering in industry — KT DS · SK Inc. AX.

Software Engineer & Manager

Back-end Server Engineer
Programming languages, frameworks & tools I work with.
Java
Python
JavaScript
C
R
Spring Boot
Spring WebFlux
FastAPI
TensorFlow
LangGraph
github.com/TAEHYUKLEE ↗
Domains I've worked in, from research to production.
Basic statistics (correlation, R², skewness), model selection via linear regression (variable selection), imputation for missing values.
Reinforcement learning (control, time forecasting) and supervised learning (time forecasting).
Finite Difference method (cavity lid-driven), spectral method (Burgers equation, channel flow).
Monitoring systems (API gateway, etc.), data collection & aggregation systems, server configuration (Web server, WAS, middleware).
AWS public-cloud architecture (VPC, public / private subnets, IGW, NAT Gateway · Instance) and self-hosted Linux servers — Nginx, NAT gateway, auto-renewing HTTPS (Let's Encrypt).
INTERESTS
Turbulence control for drag reduction through deep reinforcement learning
Reinforcement learning for skin-friction drag reduction in turbulent flow
KSME Fluid Engineering Division — 2021 Spring Conference ↗ 난류에서의 표면 마찰 저항 감소를 위한 강화학습
Forecasting fine-particulate-matter (PM2.5) concentration using a neural network
KSME 2019 Spring·Autumn Conference ↗ 신경회로망을 이용한 초미세먼지 PM2.5의 농도 예측
Engineering Contest — 2nd prize
The Korean Society of Mechanical Engineers · Conference paper
Conference Poster presentation — Award