Profile
EM simulation engineer who makes AI work in industrial software. Five-plus years of CAE application and development experience across a research institute, a Chinese CAE vendor (as team lead), and the China platinum agency of Dassault Systèmes. Career arc: simulation practice → secondary development and automation → AI-agent integration with industrial software. Proficient in CST secondary development and MCP-based CST automation; every AI-produced result ships with its verification path. Holds 5 granted invention patents (3 as first inventor) and 2 design patents; familiar with the full patent lifecycle.
Experience
Senior CST Engineer — Dassault Systèmes’ China platinum agency (Apr 2024 – present)
Support 50+ customers across consumer electronics, automotive, medical, telecom and research, resolve 40+ challenging technical cases, deliver five public technical webinars, and lead CST-AI initiatives covering MCP-based automation and knowledge-base skills.
RF Simulation Engineer / Team Lead — a Chinese CAE software vendor (Mar 2022 – Feb 2024)
Led a 5-person application team in secondary development and promotion of the domestic solver “Paiyue” and projects in full-vehicle EMC, 77 GHz radar testing, bio-electromagnetics, absorbing materials, and electrically-large chamber simulation, securing 5 invention and 2 design patents.
Patent Agent Engineer — an IP law firm (Sep 2021 – Mar 2022)
Handled search, drafting, and office-action responses for 30+ electrical invention and utility-model applications.
Assistant Engineer — a research institute (Dec 2020 – Sep 2021)
Performed forward design of 4G-band filters with EM and multiphysics simulation plus material parameter extraction and calibration, contributing to 10+ patent filings per month.
Selected Projects
- Smart connected vehicles: full-vehicle EMC (FDTD), lane-level multi-vehicle communication, 77 GHz radar darkroom selection testing (lead), Cable Studio simulation of complex harnesses — conducted properties, SI, EMI, EMS
- Bio-electromagnetics: FDTD human absorption analysis, MRI RF safety (co-authored paper), base station / mobile / charging scenarios
- Electrically large problems and hybrid solvers: full-wave simulation of a large anechoic chamber with low-RCS turntable at Gcells scale; surface-equivalence hybrid solver workflows
- Absorbing materials: EPP absorption and thermal effects; bow-tie / darkroom / RCS quantitative test simulation
- Antennas and arrays: irregular binary subarray optimization (patented), phased arrays, antenna-environment interaction
- Multiphysics: EM-thermal coupling, forward design of 4G-band filters with multiphysics simulation
- CAE software development: Paiyue secondary development, CST Python API automation, MCP-based CST automation, RAG knowledge base
- Open-source AI tooling: PatentDigest (patent parsing as a portable agent skill), information-clipper (web clippings into a structured local knowledge base), AGENTS-Minesweeper (screen-vision solver, deterministic algorithms)
- CST engineering knowledge skill: a doc-retrieval skill for CST engineering knowledge (code not yet public)
Selected Case Notes
Automated simulation delivery (client project, anonymized; method open for discussion)
One pipeline from modeling to generated report — pre-processing (auto model creation, layout import, boundary/mesh/monitor setup), three compute semantics (as-is, parameter injection, parameter override), staged post-processing (S-parameters, surface currents, far-field captures), and a generated report per run. Shipped 8 releases in 76 days, each immediately usable. A separate test harness auto-verifies every stage — logs, status codes, timing, parameters — so reliability came from batch verification, not manual review.
Solver-selection diagnosis in bio-EM thermal simulation (AI-assisted workflow)
A heating model always showed a uniform temperature field against physical expectation. The setup used a classic thermal solver where fluid flow demanded conjugate heat transfer; switching the solver produced a sensible result within minutes. The takeaway is the verification loop: outputs are fully determined by inputs, and every step is traceable — model & input review against the real scenario, comparison with experience and measurement, physical intuition as a sanity gate.
IP & Publications
- 5 granted invention patents (3 as first inventor), including CN116433794B, CN116775097B, CN117034636A, CN115566446B; 2 design patents
- Co-authored paper: Temperature Distribution of Implantable Neurostimulator Electrode Tip Structures in a 3T MRI RF Environment (with University of Shanghai for Science and Technology and Shanghai Institute of Medical Device Testing)
Talks & Content
- Invited talk, Medidata NEXT Shanghai (SIMULIA): “EM Simulation in Life Science Applications” (Sep 2025)
- Invited talk, SIMULIA EM Simulation Technology Seminar: “Python for Efficient CST Simulation” (May 2025)
- Five public technical webinars on the company’s Bilibili channel (700–2,800 views each, 2024–2025): Hybrid Solvers, Antenna Simulation Overview, Cable Studio, Array Antenna Practice, Python Automation
- Invited talk, ShanghaiTech University: “CST in Biomedical Engineering Applications” (Nov 2024)
- Online course “3D Electromagnetic and Thermal Full-wave Simulation Practice” (DISA platform)
- 2020 GitHub Archive Program Arctic Code Vault contributor
Education
Guilin University of Electronic Technology, B.Eng. in Electronic Science and Technology (EM fields and microwave techniques), 2017–2021, top 30% of the major.