Sort by
Refine Your Search
-
Category
-
Country
-
Program
-
Field
-
computing pipelines and large-scale GPU resources, to scale LLM development and deployment. Your profile PhD in machine learning, computer science, computational biology, computational neuroscience
-
methods on accelerated or heterogeneous computing platforms, including GPUs, FPGAs or other specialised hardware, for real-time or large-scale physics applications. We welcome candidates whose work is
-
). The applicant’s experience must be related to the tasks to be conducted, to the job description herein or to: · Training of multimodal models based on deep learning and GPU computing. · Evaluation and deployment
-
for government and commercial sponsors. A particular strength is real-time RF system development leveraging in-house FPGA, GPU, and SDR platforms. Importantly, SDD has successfully transitioned multiple
-
computing platforms, including Jetson AGX Orin, Jetson Thor, or comparable GPU-based robotic compute systems. Experience building, modifying, or troubleshooting Linux device drivers or integrating vendor SDKs
-
such as TensorFlow, PyTorch, or JAX is a must and experience with GPU-based experimentation and cluster computing (e.g., Docker, Slurm) a plus. In addition to the above, there is also a mandatory
-
Tech. ARC currently operates compute resources comprised of • 50,000+ CPU cores • 500+ GPUs • over 10 PB of storage • a world class immersive visualization lab. We continuously evaluate emerging research
-
solutions on modern HPC and GPU-accelerated systems. This role includes supporting climate and geophysical science applications, enabling large-scale AI training and inference workflows, and contributing
-
of computational resources (e.g., GPU servers and related infrastructure). Contributing to student and industry project work, including data preparation, modelling, evaluation, and deployment-related tasks where
-
research. What about computing power? Well, BOLD also has access to national compute resources (3M GPU hours on Isambard for the first 1.5 years) and we are working hard to get to 5000 H100 equivalent