Santa Clara, CA30+ days ago
Ways to stand out from the crowd: ā¢ Proficiency with robotics frameworks (e.g., ROS2) and NVIDIA simulation and AI platforms such as Isaac Lab, Isaac Sim or Cosmos ā¢ Experience with AI/ML training workflows and distributed job orchestration using tools like Ray ā¢ Deep expertise of transformer networks and experience deploying NVIDIA inference technologies (Dynamo, NIM, Triton, vLLM) using acceleration techniques like quantization ā¢ Experience with large scale data curation techniques and optimization ā¢ Broad technical expertise across networking, compute, and storage systems (e.g., S3, NFS, Lustre), with hands-on experience building and debugging APIs (REST, gRPC) as well as relevant certifications such as NVIDIA Certified AI Engineer, Certified Kubernetes Administrator (CKA), or Cloud Solutions Architect. ā¢ Develop a deep understanding of robotics workloads scaling and help translate those into optimal architectures for partners ā¢ Collaborate with DevOps teams to orchestrate data preprocessing, distributed training and inference workloads to optimize job scheduling, costs, storage access, and networking across hybrid and multi-cloud Kubernetes environments (e.g., AWS, Azure, GCP, on-prem) ā¢ Accelerate inference pipelines using NVIDIA NIM, TensorRT-LLM, vLLM, SGLang, and other engines to enable seamless, disaggregated inference architectures.