Jiarui Li

MEng · Electrical & Computer Engineering · Illinois, USA

BEng · Computer Science & Electronic Engineering · Liverpool, UK

• GPA: 3.98 / 4.0.
• Final-year project in indoor AR navigation and embedded sonar coursework.

BEng · Computer Science & Technology · Suzhou, China

• GPA: 3.8 / 4.0; top 3%.

Software Engineering Intern · Remote

• Deployed X-VLA locally on Linux for a vision-guided cloth-folding simulation pipeline involving multimodal observations, policy actions and cross-embodiment soft-prompt adaptation.
• Implemented client/server inference and motion-trajectory visualization; trained and evaluated on official demonstration data and compared predictions before and after training.
• Used trajectory replay, MuJoCo visualization and a basic Isaac Sim cloth scene for staged validation of action mapping, cloth physics and simulator issues.
• Adjusted cloth stiffness and configuration in the company simulation code, then packaged and handed over a reusable cloth component.

Algorithm Engineer Intern · Computer Vision · Wuhan, China

• Reproduced and fine-tuned YOLOv8, SAM and Florence-v2 on company factory and automotive data for seatbelt, helmet and smoking detection.
• Used data augmentation and confidence fusion for small and occluded objects. The resume reports accuracy increasing from 60% to 85% and an approximately 12% mAP improvement over single-model baselines; the evaluation definition remains to be clarified.
• Optimized inference with PyTorch/TensorRT and fast pre-screening. The resume reports throughput increasing from 3 FPS to 10 FPS.
• Integrated model outputs into an embedded safety-monitoring workflow, supporting local updates on company data.

Metrics are reported in the resume; evaluation definitions and team attribution remain to be clarified. See the project details.

Internship · VisionX LLC

Local deployment, inference communication, trajectory visualization, staged validation and cloth-component handover.

Final-year project · University of Liverpool

Two-app iOS architecture, walking-trajectory reconstruction, route-graph modeling and navigation integration.

Internship · VIA Technologies, Inc.

Model reproduction and fine-tuning, augmentation, confidence fusion, inference optimization and embedded integration.

Course project · University of Liverpool

Hardware and communication integration, sensor calibration and interpolation, Python visualization and mount design.