Deep Learning Engineer - LLM and VLM Model Compression

NVIDIA is a technology company working on advanced AI software and hardware. The role involves collaborating with teams across NVIDIA to develop deep learning frameworks for compressing large language and vision-language models.

• 8+ years of experience in Deep Learning and software development.
• BSc, MS, or PhD degree in Computer Science, Computer Architecture, or related field.
• Hands-on experience with LLM or VLM model training or inference.
• Excellent Python programming skills.
• Extensive knowledge of at least one deep learning framework (PyTorch required).
• Strong problem solving and analytical skills.
• Knowledge of algorithms and deep learning fundamentals.

Nice to have:

• Experience with model compression techniques such as pruning, NAS, distillation, and quantization.
• Experience building deep learning frameworks for training, inference, or model compression.
• GPU programming experience (CUDA or OpenCL).
• First-author publication in a top-tier deep learning or AI conference.

• Design and implement a deep learning framework for compressing large language and vision-language models.
• Develop and integrate new algorithms for pruning, neural architecture search (NAS), and distillation.
• Experiment with compressing the latest LLMs and VLMs and analyze their performance.
• Collaborate with researchers and engineers to improve design, usability, and performance.
• Lead best practices for building, testing, and releasing deep learning software.

The position is located in Poland with additional locations in Switzerland and remote options. NVIDIA is an equal opportunity employer and does not discriminate based on race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.