Develop large-scale supervised learning and reinforcement learning training frameworks to support multi-modal foundation models for AVs capable of running on thousands of GPUs; Optimize GPU and cluster utilization for efficient model training and fine-tuning on massive datasets;
Design, develop and optimize server application responsible for managing NVLink networks of thousands of network devices! Collaborate with multiple teams in our multi-functional environment on developing new features and improvements.
Design new hardware features for future processing architectures targeted at deep learning workloads, for both training and inference. Advance the state of parallel computation. Be knowledgeable about future parallel programming models and their impact to hardware.
Providing Ethernet and routing expertise to customers during project delivery to design, architect and test Ethernet networking AI solutions.
Develop drivers/tests and frameworks that can most efficiently stress the SOC/System and to see them through the full software development lifecycle. This can include working in user-space, kernel, firmware or boot software of SOC.
Develop infrastructure software and tools for large-scale pre-training, post-training, and inference. Develop and optimize tools and libraries to improve infrastructure efficiency and resiliency. Co-design and implement APIs for integration with NVIDIA's resiliency stacks.
Architecting and maintaining custom profiling frameworks that provide a unified view of execution across CPU (multi-core/multi-socket) and GPU (multi-node/NVLink) environments.
Showing 561–570 of 1,261 jobs
Show more jobs
Germany, DE
Canada, CA
France, FR