-
for performance, cost-efficiency, and low-latency inference Develop distributed model training and inference architectures leveraging GPU-based compute resources Implement server-less and containerized solutions
-
inference Develop distributed model training and inference architectures leveraging GPU-based compute resources Implement server-less and containerized solutions using Docker, Kubernetes, and cloud-native
-
at MGHPCC facility (i.e., data center, compute, storage, networking, and other core capabilities). Deploy, monitor, and manage CPUs, GPUs, storage, file systems, networking on HPC systems. Develop and deploy
-
conferences. Qualifications: PhD in computer science with file systems, GPU architecture experience. Proven ability to articulate research work and findings in peer-reviewed proceedings. Knowledge of systems