Yifan Wang
Research Engineer at Globee,inc
Tokyo, Japan
Summary
🤩
Rockstar🎓
Top SchoolYifan Wang is a research engineer with a decade of experience turning ML and LLM research into reliable, production-ready systems, currently building LLM-powered English learning features in Tokyo. He blends a strong foundation in computer vision, reinforcement learning, and data science with practical expertise in inference optimization, quantization, and cost-/latency-aware cloud deployments. Yifan emphasizes system behavior over surface-level generation quality, designing workflows that use retrieval, tool orchestration, validation, and explicit decision logic to handle ambiguity and improve safety. He moves prototypes to production reliably, partnering closely with product teams to translate vague requirements into predictable, auditable systems. His open-source work includes implementations of classic RL algorithms from Sutton & Barto, reflecting both theoretical grounding and hands-on coding ability. Fluent in Chinese, English, and Japanese, he prefers roles where technical judgment and measurable impact matter more than demos or hype.
10 years of coding experience
2 years of employment as a software developer
Master's Degree, Applied Data Science, Master's Degree, Applied Data Science at University of MichiganBachelor of Science (BS), Mathematics Of Computation, Bachelor of Science (BS), Mathematics Of Computation at University of California, Los Angeles
Chinese, Chinese, Japanese
Github Skills (8)
algorithm10
data-structures10
algorithms10
python10
dynamic-programming10
reinforcement-learning10
data-structure10
numpy10
Programming languages (6)
Github contributions (5)
LyWangPX/Reinforcement-Learning-2nd-Edition-by-Sutton-Exercise-Solutions
Apr 2019 - Nov 2022
Solutions of Reinforcement Learning, An Introduction
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:262 commits, 11 PRs, 254 pushes in 3 years 7 months
Contributions summary:Yifan contributed solutions to reinforcement learning exercises from "Reinforcement Learning, An Introduction." The contributions included implementations in Python, specifically for exercises related to grid world problems and the development of a dynamic programming solution for a car rental problem. Furthermore, the user demonstrated skills in implementing value iteration and policy improvement algorithms, critical for the book's reinforcement learning concepts. The code showcased an understanding of key reinforcement learning concepts like value functions, policies, and state-action spaces.
exercise-solutionsself-studyreinforcement-learningdeep-reinforcement-learningreinforcement
LyWangPX/Reinforcement_Learning_Coding_Examples
Mar 2020 - May 2020
Contributions:57 commits, 30 pushes in 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.