MIT 18.06 · Linear Algebra
Building the linear-algebra and matrix-method foundation.
ABOUT / TRAJECTORY
About
From geotechnical research and offshore wind engineering to deep RL and neuromusculoskeletal modeling.
My core training is in mechanics, computational modeling, and engineering analysis. After completing an MSc in geotechnical engineering at Central South University and working on offshore foundations at Goldwind, I began systematically building the machine-learning and reinforcement-learning foundations needed to study human movement and neuromusculoskeletal control.
Education
Central South University
Sep 2017–Jun 2024
- MSc
- Geotechnical Engineering · Sep 2021–Jun 2024 · Recommended admission · Rank 6/149
- BEng
- Urban Underground Space Engineering · Sep 2017–Jun 2021 · Rank 7/89
- Recipient of the National Scholarship.
Research
Geotechnical & offshore research
Central South University · 2020–2024
- Focus
- Pile–soil interaction, lateral response, and plastic development in reinforced-concrete piles
- Methods
- ABAQUS, UMAT, MATLAB, and analytical models
- Output
- First-author paper published in Ocean Engineering
Experience
Goldwind
Jul 2024–Jul 2025
- Department
- R&D Center · Offshore Structural Technology
- Role
- Offshore Foundation Design Engineer
- Supported fixed offshore wind foundation work during pre-bid and bidding stages.
- Focused on geotechnical parameter interpretation, pile–soil interaction modeling, and engineering-calculation automation.
- Used SACS, PLAXIS, and MATLAB for parametric monopile analysis, finite-element calibration of p–y curves, and fatigue checks.
- Contributed to internal geotechnical design guidance and platform development.
Current
CS285 · Deep Reinforcement Learning
In progress
- Focus
- Core algorithm derivations, implementations, and experiments
- Next milestone
- Reproduce core models and training results from the MS-Human-700 series
- Progress
- Planned after CS285; not started yet
Learning path
A progression of capabilities, not a wall of course certificates.
Andrew Ng · Machine Learning
Establishing the basic framework of ML problems, models, and optimization.
Karpathy · Neural Networks: Zero to Hero
Implementing neural networks from scratch to understand how training works.
Mu Li · Dive into Deep Learning (D2L)
Studying deep-learning models systematically through code and experiments.
Shiyu Zhao · Mathematical Foundations of Reinforcement Learning
Understanding fundamental RL problems and classical algorithms mathematically.
Sutton & Barto · Reinforcement Learning: An Introduction
Building a unified view of classical reinforcement-learning methods.
CS285 · Deep Reinforcement Learning
Current stage: core deep-RL algorithms, derivations, and implementations.
MS-Human-700 · Reproduction milestone
After CS285, reproduce core models and training results from the MS-Human-700 series.
Current toolkit
This combines tools used in earlier research with current areas of study; it is not a proficiency ranking.
ABAQUS / UMATSACSPLAXISMATLABSANISANDCDPPythonDeep RL
Stay in touch
If you are interested in deep reinforcement learning, neuromusculoskeletal modeling, or cross-domain computational research, feel free to connect on GitHub.