Zhilin Xu

Engineered agentic AI systems and RAG pipelines for real-time root-cause analysis, reducing diagnostic latency and delivering fault-tolerant, low-latency observability insights at enterprise scale.

Implemented graph-optimization and INT8/FP16 quantization, designed runtime scheduling for NPU–CPU workloads under QNX RTOS, and extended toolchain importers to optimize edge inference throughput.

Designed and developed accelerator-specific ML operators and automated testing/validation pipelines to optimize model training and inference for Cornami's low-power AI chip.

Automated wafer defect matching and visualization pipelines, improving defect matching efficiency by 15% and delivering actionable insights for Surfscan systems.

Built robot communication infrastructure and cloud/Android interfaces, implemented real-time voice integration and path-planning algorithms to support multi-robot control.

Integrated Orion compare widget into Rational Team Concert code-review web app, resolved cross-stream merge discrepancies, and fixed 100+ bugs within two weeks.

Updated Android game UI and implemented matchmaking logic, contributing to a 4.5 Google Play rating for Word Streak and improving user socialization.

Developed web and Android applications to register handwritten forms via digital pens, integrating device data capture with backend services.

Supported database administration tasks and data operations to maintain service reliability and support internal applications.

Contributed to software engineering projects supporting TELUS services, assisting with development and testing tasks.

Completed a Master’s degree in Computer Science with a data science focus, emphasizing machine learning, data analysis, and applied AI techniques.

Earned a Bachelor’s degree in Computer Engineering with coursework and projects covering embedded systems, algorithms, and software engineering.