Deep Learning Engineer - LLM and VLM Model Compression

We are looking for DL engineers passionate about building deep learning frameworks for large language (LLM) and vision language (VLM) model compression that push the boundaries of AI efficiency. In this role, you’ll collaborate with world-class teams across NVIDIA to advance both the software and hardware stack that powers modern AI.

Join the team building software used by the entire world. Work with world class engineers and researchers to build next-generation deep learning frameworks for compressing LLM and VLM models through pruning, distillation, and neural architecture search (NAS). Work on most powerful, enterprise-grade GPU clusters capable of hundreds of Peta FLOPS and on unreleased hardware before anyone in the world. Are you ready for this challenge?

What you’ll be doing:

• Design and implement a deep learning framework for compressing large language and vision-language models to deliver highly optimized, high-performance AI systems used worldwide.

• Develop and integrate new algorithms for pruning, NAS, and distillation in collaboration with NVIDIA researchers and engineers.

• Experiment with compressing the latest LLMs and VLMs, analyzing their performance and behavior across diverse workloads.

• Collaborate with researchers and engineers across NVIDIA, providing guidance on improving the design, usability and performance of workloads.

• Lead best-practices for building, testing, and releasing DL software.

What we need to see:

• 8+ years of experience in Deep Learning and SW Development.

• BSc, MS or PhD degree in Computer Science, Computer Architecture or related technical field.

• Hands-on experience with LLM or VLM model training or inference.

• Excellent Python programming skills.

• Extensive knowledge of at least one DL Framework (PyTorch, TensorFlow, JAX, MxNet) with practical experience in PyTorch required.

• Strong problem solving and analytical skills.

• Algorithms and DL fundamentals.

Ways to stand out from the crowd:

• Experience applying and implementing model compression techniques such as pruning, NAS, distillation, and quantization.

• Experience building deep learning frameworks for training, inference, model compression, or related topic.

• GPU programming experience (CUDA or OpenCL) is a plus but not required.

• First-author publication in a top-tier deep learning or AI conference.

NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most brilliant and forward-thinking people in the world working for us. If you're creative and autonomous, we want to hear from you! We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.