NVIDIA izstrādā paātrinātās skaitļošanas un mākslīgā intelekta tehnoloģijas, ko izmanto spēļu, datu centru, mākoņdatošanas, veselības aprūpes, ražošanas, autobūves un robotikas jomās. Uzņēmuma darbība aptver grafikas procesorus, mākslīgā intelekta platformas, simulāciju un nozarei paredzētus risinājumus, tostarp NVIDIA Omniverse sadarbības 3D darbplūsmām, NVIDIA DRIVE autonomo transportlīdzekļu izstrādei un NVIDIA Clara veselības aprūpes lietojumprogrammām. Uzņēmums apvieno lielas tehniskās komandas, kas strādā pie infrastruktūras un programmatūras, kura nodrošina progresīvu skaitļošanu, simulāciju un mākslīgā intelekta vadītu analītiku.
Senior Deep Learning Software Engineer, Inference
Build and tune GPU-accelerated software for large-scale AI inference at NVIDIA. Improve open-source serving frameworks and model performance across NVIDIA accelerators.
Apraksts
- Build and optimize GPU-accelerated software for AI applications.
- Develop and maintain open-source frameworks for large-scale model serving and inference.
- Enhance inference platforms and support language model deployment and serving.
- Implement new algorithms for release through inference frameworks.
- Drive performance gains for advanced LLM and generative AI models across NVIDIA accelerators.
- Apply tools such as CUTLASS, OAI Triton, NCCL, and CUDA kernels to optimize model-serving pipelines.
- Analyze and tune deep learning models in LLM, multimodal, and generative AI.
- Scale model performance across architectures and NVIDIA accelerator types.
- Contribute features and code to NVIDIA inference libraries, vLLM, SGLang, FlashInfer, and LLM software.
- Work with teams across frameworks, NVIDIA libraries, and inference optimization solutions.
Prasības
- Master’s degree, PhD, or equivalent experience in a relevant field such as computer engineering, computer science, EECS, or AI.
- At least five years of relevant software development experience.
- Strong C/C++ programming and software design skills.
- Agile software development experience is helpful.
- Python experience is a plus.
- Production experience training, deploying, or optimizing deep learning model inference is a plus.
- Experience in performance modeling, profiling, debugging, and code optimization, or knowledge of CPU and GPU architecture, is a plus.
- GPU programming experience with CUDA, OAI Triton, or CUTLASS is a plus.
- Experience contributing to deep learning software projects such as PyTorch, vLLM, or SGLang.
- Experience with multi-GPU communications, including NCCL or NVSHMEM.
Priekšrocības
- Comprehensive benefits package.
- Equity.