MSc student specialising in AI infrastructure, machine learning, and scalable computer vision systems. Previously a Software Engineer Intern at Taiwan Mobile, building media-rich customer experiences and intelligent tooling.
I build modular deep learning pipelines and full-stack analytics prototypes that bridge research and production—spanning automated basketball video analysis, one-shot learning for classroom roll call, and cloud-native ML products.
AI infrastructureComputer VisionCloud MLMLOpsPrototyping
Human-in-the-loop ML
Scalable CV
About
AI & CV
MSc Advanced Computing @ King’s College London. I build modular deep learning pipelines and full-stack analytics prototypes that bridge research and production.
Undergraduate Research Assistant (NSTC Funded) Jul 2023 – Feb 2024
Led one-shot face recognition research for classroom roll call, combining data augmentation, transfer learning, and triplet networks to raise accuracy by up to 20%.
Built TensorFlow training pipelines with ensemble methods such as SimCLR and Triplet Networks for stability across limited datasets.
Co-authored the TAAI 2023 paper and received the NSTC Undergraduate Research Grant (<5% acceptance rate).
Chang Ching Yu Memorial Library
Taoyuan, Taiwan
Library IT Intern Feb 2022 – Feb 2024
Enhanced a Python-based thesis checker with Selenium and pdfplumber automation, reducing manual review and improving document validation accuracy.
Automated BibTeX-to-CSV workflows to streamline citation data processing for academic databases.
Provided technical support for ~500 staff and students, improving software access and troubleshooting turnaround.
Taiwan Mobile Co., Ltd.
Taipei, Taiwan
Software Engineer Intern Jul 2023 – Aug 2023
Developed a Java-based batch image conversion tool to standardise formats and reduce manual preprocessing for design teams.
Prototyped HTML5/WebRTC media capture flows with Vue.js and jQuery, enabling more efficient customer service interactions.
Explored YOLOv8-driven logo detection to identify outdated assets across web properties and support dataset curation.
Research
Basketball Game Visualization and Performance Evaluation
Yu-Han Lee Feb 2025 – Jul 2025
Built my MSc thesis project under the supervision of faculty at King's College London.
Engineered a modular CV pipeline combining YOLOv8 detection, ByteTrack multi-object tracking, R(2+1)D action recognition, CLIP team identification, and homography-based tactical mapping.
Deployed training and inference workflows on an HPC cluster with PyTorch, achieving YOLOv8 mAP 0.97 and R(2+1)D accuracy of 86%, and generated player trajectories, speed profiles, and play insights.
Classroom Roll Call Technology Based on One-shot Learning Yu-Han Lee, Wen-Jun Lin, Wen-Hsin Hsu Sep 2022 – Sep 2023 NSTC Undergraduate Research Grant · TAAI 2023
Investigated few-shot learning for classroom roll call using deep learning to overcome limited training images per student.
Combined data augmentation, transfer learning, and triplet-network training to reach reliable face recognition with a single enrolment photo per student, culminating in a TAAI 2023 publication.
Education
King's College London
London, United Kingdom
MSc Advanced Computing Sep 2024 – Sep 2025
Coursework highlights: AI Planning, Multi-Agent Systems, Computer Vision, Security Engineering, Big Data Technologies, Distributed Ledgers & Crypto-Currencies, Pattern Recognition & Deep Learning, FinTech Software Engineering.
Thesis: “Towards Automated Basketball Video Analysis through Modular Deep Learning Architectures: Detection, Recognition, and Tactical Insight.”
Chung Yuan Christian University
Taoyuan, Taiwan
BSc Information and Computer Engineering Sep 2020 – Jan 2024
GPA 3.93/4.0 (Dean's List, ranked 2nd/56 in Spring 2020). Completed core modules in data structures & algorithms, operating systems, computer organisation, system programming, networks, probability & statistics, and data science.
Year 1 completed in Commercial Design (Product Design Division) prior to transferring into Information and Computer Engineering.
Projects
Payment Transaction Simulator Yu-Han Lee Oct 2025
Built a card-network payment authorisation simulator with FastAPI, SQLite, and Docker, covering analysis, design, testing, and deployment.
Delivered RESTful endpoints for payments, transaction lookups, statistics, and resets, backed by fraud heuristics, validation layers, and Swagger-driven demos to visualise the flow.
Cloudflare AI Chat Yu-Han Lee Sep 2025
Built a multi-model Workers AI chat experience with Durable Object-backed conversation state, Vectorize-powered document grounding, and file uploads for RAG-style responses.
Added streaming responses, model selection, and tracing to make the workflow production-friendly on Cloudflare’s edge stack.
FusionGuard Analytics Yu-Han Lee Aug 2025
Credit-card fraud and churn analytics platform with Spark feature pipelines, MLflow experiment tracking, FastAPI agent endpoints, and Streamlit monitoring dashboards.
Delivered explainable model outputs, cohort filters, and operational tooling to help analysts iterate quickly.
Our-Scheme (Programming Language Project) Yu-Han Lee Feb 2023 – Jul 2023
Built a custom C++ interpreter for evaluating Scheme-like expressions based on formal language specifications.
Implemented lexical analysis, syntax parsing, and evaluation modules to provide a modular architecture for handling user-defined programs.
5-stage Pipelined CPU (Computer Organization Project) Yu-Han Lee Feb 2022 – Jul 2022
Designed and implemented a five-stage pipelined MIPS-Lite CPU using Verilog HDL and ModelSim, covering arithmetic, logic, memory, and branching operations.
Verified control and data hazards through waveform simulations and test benches to ensure correct execution.
SIC/SICXE Assembler (System Programming Project) Yu-Han Lee Aug 2021 – Jan 2022
Developed a two-pass assembler in C++ for SIC and SIC/XE machine languages, translating source code into object code with syntax validation and instruction set support.
Certifications
Oracle Cloud Infrastructure 2025 Certified Data Science Professional (1Z0-1110-25) — Oct 2025
Oracle Cloud Infrastructure 2025 Certified Generative AI Professional (1Z0-1127-25) — Sep 2025
Oracle AI Vector Search Certified Professional (1Z0-184-25) — Sep 2025
