Zhijian Fan
Education
Beijing University of Posts and Telecommunications (BUPT) | Queen Mary University of London
2023 — 2027 ExpectedB.Eng, Telecommunication Engineering and Management
Experience
Faculty of Science, The University of Hong KongResearch Project Member - Substrate-Level Failure Prediction
2025 — Present- Joined a faculty research project on semiconductor manufacturing analytics and substrate-level failure prediction.
- Translated sparse wafer defect patterns into modeling objectives and experiment plans for predictive analysis.
- Contributed to iterative research discussions on model robustness across different wafer specifications and manufacturing conditions.
International School, Beijing University of Posts and TelecommunicationsDeveloper - Student Information Management and Personal Planning System
2025 — Present- Helped build and launch a student data platform that reached 2,000+ users after school-wide promotion.
- Owned end-to-end product iteration across user workflows, data analysis logic, backend services, and deployment reliability.
- Worked on a data flywheel loop that connected real user activity with continuous improvements to prediction and recommendation features.
Innovation Centre, Queen Mary University of LondonDeveloper - QMUL WinterHack Challenge
2026 — 2026- Developed a ROS2-based robotic arm vehicle on Raspberry Pi, focusing on software and algorithm design.
- Implemented SLAM for localization and mapping and A* search for path planning.
- Integrated perception and control modules to enable autonomous navigation and object grasping.
OxCam Programme, University of CambridgeParticipant - Large Language Model & Generative AI Summer Camp
2024 — 2024- Completed three weeks of coursework and interdisciplinary research under the supervision of Prof. Pietro Lio.
- Explored generative AI and computer vision research directions through a final project on sign language recognition.
- Delivered the final presentation and received a Distinction grade.
Projects
Substrate-Level Failure Prediction Based on Deep LearningData Science / Deep Learning Research
2025 — Present- Prepared substrate-level defect data for modeling, including sparse-pattern preprocessing and feature extraction.
- Built neural network experiments to predict die failure from manufacturing defect signals.
- Evaluated model behavior across wafers with different specifications to improve robustness and generalization.
- Connected prediction outputs with production-efficiency goals in the semiconductor manufacturing context.
Student Information Management and Personal Planning SystemAI / Data Platform
2025 — Present- Modeled long-term learning and career trajectories with a Transformer-based framework on student data.
- Designed recommendation and prediction logic for personal planning scenarios.
- Implemented lightweight database, relay, and authentication services to support cross-network usage.
- Built feedback pathways from user activity to platform iteration, supporting a data flywheel after launch.
QMUL WinterHack ChallengeRobotics / Embedded AI Project
2026 — 2026- Built a ROS2-based robotic arm vehicle on Raspberry Pi with a focus on software and algorithm design.
- Implemented SLAM for localization and mapping, plus A* search for path planning.
- Integrated perception and control modules to enable autonomous navigation and object grasping.
Self-Supervised Hand Pose Estimation and RecognitionGenerative AI / Computer Vision Research
2024 — 2024Distinction
- Studied self-supervised learning methods for hand pose estimation.
- Explored how hand-pose representations can support sign language recognition.
- Produced the final presentation titled "Self-Supervised Hand Pose Estimation and Recognition" and received a Distinction grade.
Skills
Programming:
Python, PyTorch, C
AI & Data:
Machine Learning, Deep Learning, Transformer Models, Predictive Modeling, Data Analysis
Robotics & Algorithms:
ROS2, SLAM, A* Path Planning, Raspberry Pi
Backend & Systems:
Backend Services, Database Services, Authentication Services, API Integration, Lightweight Backend Architecture
Languages:
Chinese (Native), English (IELTS 7.0)
Coursework:
Machine Learning and Data Analysis, Data Structures, Signals and Systems Theory, Digital Signal Processing, Digital Circuit Design, Electronic and Circuit Foundation
Interests:
AI4S, Agentic AI