Zhilin Xu

I am an AI/ML engineer with 7+ years designing high-performance, production-grade AI systems across cloud, embedded, and enterprise platforms. I specialize in model optimization, quantization, mixed-precision training, and hardware-aware acceleration to deliver measurable latency and cost improvements.

I have built agentic orchestration frameworks, AI-as-a-Service infrastructures, and RCA pipelines integrating LLM reasoning, telemetry, and vectorized knowledge graphs for enterprise observability. At AppDynamics I led an AI-Assisted Root-Cause-Analysis initiative and deployed multi-agent reasoning and telemetry-driven inference on Kubernetes and AWS.

My work spans embedded SoC optimization, compiler and kernel collaboration, and MLOps best practices—enabling seamless deployment from research to production. I am passionate about bridging AI research and systems engineering to create reliable, interpretable, and high-throughput AI solutions that drive real-world business impact.