Nvidia Senior Performance Engineer - Deep Learning 

United States, Texas 

579053404

What you will be doing:

• Develop and optimize open-source libraries, like Transformer Engine, which enables the fastest training of Large Language Models using low precision data formats, and TensorFlow Distributed Embeddings, providing ability to easily scale training of huge recommender systems on multiple GPUs.

• Study and tune Deep Learning training workloads at large scale, including important enterprise models.

• Build and support NVIDIA submissions to community benchmarks like MLPerf.

• Optimize the performance of influential, modern Deep Learning models coming out of academic and industry research, for NVIDIA GPUs and systems.

• Explore new technologies and advise design of new hardware generations and core platform software components.

What we need to see:

• BS in Computer Science, Electrical Engineering or a related field (or equivalent experience).

• Demonstrated ability with 6+ years of C++ and Python programming.

• Strong background with parallel programming, preferably on GPUs.

• Knowledge of Computer Architecture and/or Operating Systems.

• Proven experience developing large software projects.

• Excellent verbal and written communication skills.

Ways to stand out from the crowd:

• Experience with Deep Learning Frameworks, like PyTorch, JAX, Tensorflow or MXNet.

• Experience training language models.

• Background with performance analysis and profiling of workloads.

• Participation in the open-source community.

• Proven experience working with multidisciplinary teams.