Name:Jianming Tong

Bio:

I’m a computer architect focusing on improving the efficiency of AI system and privacy-preserving AI system via full-stack optimizations spanning model, system, compiler, and hardware. A few representative highlights:

1. CROSS Compiler (HPCA’26, MLSys’24 with Google): Converts Homomorphic Encryption workloads into AI workloads to be executed efficiently on TPUs — enabling immediate, scalable, and low-cost privacy-preserving AI.
2. Reconfigurable Accelerator (ISCA’24): Proposes dataflow-layout co-switching for dynamic AI workloads to sustain high throughput under runtime irregularity.

These works are deployed in NVIDIA (NV Labs) and Google (Jaxite), and recognized by Qualcomm Innovation Fellowship, Machine Learning and System Rising Star, and GT NEXT Award.

Country:United States

Affiliation:Georgia Institute of Technology

Research interests:'m a Computer Architect, focusing on system for AI and Cryptography, e.g., enabling today’s AI systems to work in a privacy-preserving manner without sacrificing performance.