Mason Jiang

Drove end-to-end engineering strategy across AI model deployment, compiler stack, and hardware-software integration. Oversaw teams spanning over 100 engineers across ML, runtime, and systems domains. Led the launch of a commercial AI platform supporting multi-billion parameter models with automated scaling and fine-tuning pipelines. Negotiated and executed co-development agreements with federal agencies and Fortune 100 enterprises to accelerate adoption of AI inference at scale.

Led the design and optimization of large-scale ML workloads on SambaNova’s dataflow-based architecture. Directed initiatives for training and deploying transformer and BERT-style models on custom silicon, enabling 10× throughput improvement in NLP inference for enterprise clients. Collaborated closely with

Led a team of data scientists and engineers to develop predictive models for system diagnostics and customer analytics. Deployed machine learning solutions that reduced support ticket volume by 18% and improved system uptime forecasting accuracy across enterprise storage products.

Responsible for system architecture design and core function development, optimizing system performance and stability; participating in technical solution formulation, code review and technical guidance, and promoting high-quality project delivery.

Conducted applied research on distributed storage and high-performance computing systems. Prototyped scalable memory and interconnect architectures for data center use, with publications and patent filings in the areas of parallel computing and system-level optimization.

Designed and implemented key modules for Citrix’s core virtualization platform, focusing on cross-platform hypervisor integration and resource management optimizations. Enhanced scalability and tenant isolation in multi-VM deployments, contributing to improved product stability for enterprise customers.