Principal Software Engineer, MLOps Infrastructure

About the Role

We are looking for a Principal Software Engineer with deep learning leadership experience to join our group! Come join NVIDIA's AI/AV Infrastructure team to develop state-of-the-art MLOps infrastructure for our advanced autonomous driving platform. Together, we will help advance NVIDIA's capacity to build and deploy leading solutions for a broad range of AI-based applications.

To ensure that our fleet of autonomous vehicles scalably record, analyze, and train our state-of-the-art machine learning models, we need to stay a step ahead of our engineering partners. We need someone who has built systems that handle petabyte-scale datasets, or is able to map their prior experience to develop systems that handle such scale. We are either directly involved or in collaboration with our partner teams in all the data lifecycle activities. To ensure we deliver on the expectations of our customers, our solutions must be scalable and performant. Our approach must embrace engineering and operational excellence practices across the stack. We need someone who can ensure that the primitives which make our platform build, scale, and ship reliably. You will embed in one of our engineering teams that has high visibility and work on both product-focused and infrastructure roles.

What You'll Be Doing

• Contributing to the development of Deep Learning software infrastructure for large scale image and video processing tasks and leading major technical projects for the team. The range of applications you'll work on includes automotive driver assistance, autonomous navigation, and robotics.
• Be a part of a dynamic product and customer-oriented team. Your expertise, creativity and leadership will help bring the future of self-driving cars to our doorsteps.
• Be responsible for data modeling, schema design, dataset curation, search, and discovery.

What We Need To See

• 10+ years of relevant work experience in high performance/distributed computing, including technical leadership experience.
• 3+ years of relevant experience in productizing deep learning systems.
• You have a BS or MS in Computer Science, Electrical Engineering, or equivalent experience.
• Experience with MLOps platforms such as Flyte, MLFlow, or similar.
• Experience with any of the deep learning frameworks: PyTorch, Tensorflow, Keras, or similar.
• Proven track record to architect and develop data-first, production machine learning pipelines that scale out to very large datasets.
• Track record of elevating teams and increasing their velocity by helping others.
• Experience mentoring junior developers.
• Good understanding of highly parallel compute, storage, and software architectures.
• Experience working with multi-node/distributed training, data loading and image/video processing.
• Excellent communication and organization skills.
• Self-motivation, outstanding collaboration with peers and users, customer focused mindset.

Ways To Stand Out Of The Crowd

• Previous experience of scaling up and optimizing HPC, computer vision or deep learning training pipelines to terabyte scale datasets is a huge plus.
• Experience with Go Lang, C++ and CUDA.
• Contributions to open-source DNN frameworks.
• Prior experience using machine learning or neural networks to tackle problems in image or video analysis.