Join NVIDIA as a Deep Learning Compiler Engineer

Company: NVIDIA

Position: Software Engineer, PyTorch - Deep Learning

Location: Open for Global Talent

Are you ready to innovate at the intersection of AI and software engineering? NVIDIA is seeking skilled Deep Learning Compiler Engineers to create state-of-the-art tools enhancing the capabilities of AI engineers worldwide. Embrace the opportunity to impact AI design, development, and deployment on a global scale, utilizing thousands of GPUs. Join our dedicated team focused on delivering perpetual excellence in software experiences on NVIDIA's leading hardware with PyTorch.

What You'll Do:

As a pivotal member of our team, your role will extend beyond traditional boundaries, driving the evolution of distributed deep learning training workloads via advanced compiler technologies. Your core responsibilities will include:

• Performing in-depth analysis of deep learning workloads to pinpoint and resolve performance bottlenecks, functional anomalies, and system inefficiencies.
• Linking performance issues directly to compiler defects or untapped optimization potentials and formulating robust resolutions.
• Working collaboratively to refine program transformations or innovate new ones following comprehensive performance reviews.
• Keeping pace with the latest developments in deep learning compilers to propose cutting-edge solutions that enhance deep learning framework efficiency.
• Conducting thorough testing and validations to ensure optimal performance and quality in model training.

What We Need to See:

Come equipped with:

• A Bachelor's, Master's, or Ph.D. in Computer Science or a related technical field, or comparable experience.
• Proficiency in Python and familiarity with machine learning frameworks like PyTorch or JAX.
• A solid understanding of compiler concepts including abstract interpretation, code representations such as SSA form and AST, code generation, and program transformations.
• Proven track record in developing sizable software projects.
• Strong capabilities in both verbal and written communication.

Stand Out With:

• Prior contributions to open-source deep learning compiler projects like TVM, or involvement in deep learning frameworks.
• In-depth knowledge of PyTorch and/or JAX internals.
• Expertise in distributed systems, parallel computing, and CUDA programming.
• Active participation in the open-source community.Experience in collaborating with diverse, multidisciplinary teams.

Ready to take your career to the forefront of technological innovation? Apply now to become part of NVIDIA's mission to pioneer a future where AI and deep learning compilers come together to create unprecedented applications and experiences.