Sort by
Refine Your Search
-
Listed
-
Category
-
Country
-
Program
-
Field
-
decays and related physics object performance studies , development of the real-time analysis (RTA) in particular with ML/AI reconstruction on hybrid GPU/FPGA architecture for the electromagnetic
-
of today’s heterogeneous hardware (multicore CPUs, GPUs, SmartNICs, disaggregated datacenters). We explore: SmartNICs & P4 switches for offloading intelligence from hosts Device-to-device communication
-
projects at CASS. The center fellows will have access to a 70,000-core Infiniband Cluster (Jubail) dedicated to the science division, several GPU-based clusters at NYUAD, and other supercomputer facilities
-
with edge computing or embedded systems (e.g., NVIDIA Jetson, Raspberry Pi) Background in real-time processing and GPU acceleration (CUDA) Participation in relevant competitions (e.g., Kaggle, computer
-
(3–4 years) Salary per university collective agreement (€39,005.40 gross/year, 30 hours/week) Access to a modern GPU cluster Conference travel and active support towards publications How to apply Email
-
optimization layers Increase inference efficiency (e.g., GPU acceleration) and assess the applicability domain of learned algorithms Publish and present your results in peer-reviewed journals and at
-
hardware architectures (multicore, GPUs, FPGAs, and distributed machines). In order to have the best performance (fastest execution) for a given Tiramisu program, many code optimizations should be applied
-
machine learning techniques, and GPU programming. The simulation results will be compared to observational data obtained using facilities worldwide including ESO and NOT. Who we are looking for A successful
-
The University of North Carolina at Chapel Hill | Chapel Hill, North Carolina | United States | about 2 months ago
GPU accelerated pipelines. In collaboration with a worldwide network of real-time data release and processing centers, the Data Access Engineer will take the alert distribution system to production
-
. Training LLMs, large-scale deep learning systems, and/or large foundation models using GPU/TPU parallelization while setting up the environment/system network under various constraints, such as limited