Tsinghua University
Engineering Management, Department of Industrial Engineering, 2025.09 - Present;GPA 3.9 / 4.0;core courses include Operations Research, Strategic Management, Systems Engineering, HR Management, Machine Learning
Personal Knowledge Studio
Yicheng Li / MercerIndustrial AI Portfolio李奕成 Mercer｜工业 AI 作品集
From precision MRI manufacturing to process engineering, automation testing, and AI product practice.
Moving toward AI Product and Industrial AI Project roles, with a focus on bringing AI into production, quality, and engineering delivery workflows.
About / About Me
I am Yicheng Li, also known as Mercer — a process and automation engineer who grew from the front line of precision MRI manufacturing. My long-term direction is to combine manufacturing know-how, AI tool capabilities, and project management methods to drive AI implementation in real industrial scenarios.
Process EngineerAutomation EngineerAI PractitionerFluent Spoken EnglishLogical ThinkingFirst Principles
Education / 教育背景
Shenzhen University
Mechanical Design, Manufacturing and Automation, Robotics Track, Bachelor, 2018.09 - 2022.06;core courses include Mechanical Design, Machine Vision, Machine Learning, Embedded Systems
OTH Regensburg
Department of Mechanical Engineering, International Exchange Student, 2019.03 - 2019.04
Working Experience / Engineering Practice
Siemens HealthineersI built my working experience in real manufacturing environments: new product introduction, process development, troubleshooting, equipment failures, quality issues, and supplier management. I believe the value of AI in manufacturing must return to efficiency, quality, stability, and deliverable outcomes. Four years at Siemens Healthineers strengthened my engineering experience and my reliability as a professional.
- Process engineer in Siemens Healthineers MR business
- Managed process, equipment, quality, and material issues across 5 mass-production lines
- Responsible for product yield above 99.8% in 2025
- MR sub-component process development / NPI
- Large magnetic field test system introduction, installation, and validation
- Advanced equipment / precision product manufacturing and validation know-how
Automation & Data / 自动化与数据
I focus on making better use of production data: from data analysis to process early warning and quality improvement. Data is not just a dashboard. It is the input to machine learning models.
- Automated testing solutions
- LabVIEW / Python / MATLAB
- Historical process data
- Test results and defect records
- Analysis reports
AI Projects / Applied Work
01
DemoLabVIEW-Based Automated Test Platform
- Context
- The existing manual testing process required multiple instruments, involved complex procedures, and took a long time to complete
- My Role
- Solution design and implementation
- Approach
- Developed a LabVIEW-based one-button test platform and built an XGBoost-based risk warning model using historical process data, test results, and defect records.
- Tools
- LabVIEW, Python, XGBoost, Manufacturing Data
- Outcome
- Saved working hours, improved testing efficiency in the manufacturing process by 90%, and achieved around CNY 400K annual cost saving
Automation TestingXGBoostQuality Prediction
Click to flipACCESS NOTE
Demo available on request
Because this platform involves real manufacturing workflows and testing logic, a demo can be shared separately after direct communication.
Click to flip02
Case LinkMachine Vision Defect Detection for Roll-to-roll Flexible Circuit Boards
- Context
- Manual defect inspection was inefficient and prone to missing defects
- My Role
- Machine vision engineering intern, involved in algorithm development
- Approach
- Used OpenCV for image processing, ROI extraction, and post-processing; built a ResNet-18 model for defect classification.
- Tools
- OpenCV, ResNet-18, Industrial Camera, Image Processing
- Outcome
- Reduced missed detection rate by 90%, improved processing speed by 50%, and achieved 98%+ recall
Machine VisionOpenCVIndustrial Inspection
Open case link03
VideoAI-assisted CT Image Diagnosis for Rare Disease
- Context
- CT lesion segmentation, recognition, and grading for rare disease diagnosis
- My Role
- Core algorithm engineer
- Approach
- Developed a Transformer-enhanced Unet-8 model for lesion segmentation.
- Tools
- Python, Transformer, Unet, Medical Image Segmentation
- Outcome
- Achieved 91.7% segmentation accuracy; trialed at Tsinghua Changgung Hospital and won first prize in the course project
Medical AICT SegmentationTransformer
Click to flipACCESS NOTE
Video available on request
The project video can be shared separately after direct communication.
Click to flipAI Agent Lab / Product Practice
I am turning AI into practical workflows: understanding business needs deeply, designing product features, quickly validating prototypes, and judging whether they are actually used and able to create value.
AI WorkflowRAGEnterprise Knowledge Base
CENTERPIECE PROJECT
Enterprise Knowledge Base Built with RAG
A real demo of an enterprise knowledge base: select private documents/data, ask questions, visualize the RAG (retrieval-augmented generation) workflow, and see how the generated context and prompt are passed into an LLM.
01Document Library
02Chunk & Embed
03Retrieve & Rerank
04LLM Context
Capability Map / 能力地图
Manufacturing Know-how
Process / Equipment / Quality / NPI / Medical Device Validation
Automation & Data
LabVIEW / Python / MATLAB / Automated Testing / XGBoost
AI & Computer Vision
OpenCV / ResNet-18 / Transformer / Unet / AI Agent
Common skills
Communication / Project Management / Spoken English / Basic German
Beyond Work / Personal Layer
I’m a person with diverse interests who loves to share and is full of energy. Outside of work, I enjoy watching movies, reading novels, and hiking, and I’m also a fan of science fiction and music. I believe in the power of connecting the dots and am committed to a long-term perspective.
Recent quoteShowing up is 80 percent of success.
Film Picks
Three films I would recommend
- Interstellar
- The Truman Show
- Still Life
Career Direction / 职业方向
My next step is to move from the manufacturing site toward AI Product Manager or industrial AI Project Manager roles. I hope to work on AI projects that truly enter business workflows, improve quality, increase efficiency, reduce risk, and create practical business value.
AI Product ManagerAI Project ManagerIndustrial Digitalization PMAI Presales / Solution EngineerFDE / Forward Deployed Engineer
Contact / 联系方式
I am open to opportunities in AI product management, industrial AI project management, manufacturing digitalization, and automation solutions. I am especially interested in applying AI to real manufacturing workflows and medical device production systems.