Senior Machine Learning Engineer, Quantized Training

at  NVIDIA

WE ARE NOW LOOKING FOR A SENIOR MACHINE LEARNING ENGINEER FOR QUANTIZED TRAINING.

NVIDIA is seeking machine learning engineers to support next-generation recipes for mixed-precision training. In this role you will (1) distill LLM research literature into its core, (2) translate literature into experiments at scale, (3) create insights to support or refute the efficacy of a technique, and (4) generate reproducible training recipes.

WHAT WE NEED TO SEE:

• Experience with PyTorch or similar frameworks such as jax/xla/etc
• Proficient in the math of machine learning
• Familiarity with FP8 for training
• Published research or significant contributions to the field of AI, particularly in algorithm development for hardware-software co-design
• PhD, M.S. degree or equivalent experience in Computer Science or a related field
• 5+ YoE working in ML / AI
• Strong written and oral communication skills
• Strong programming skills and ability to debug ML systems

• Review state-of-the-art literature in quantized training
• Build robust, reproducible, and portable training recipes
• Provide engineering support to customers using HW and SW approaches
• Collaborate closely with hardware, software, and research teams to assess and adopt deep learning algorithmic advancements in quantization
• Work with production SW teams to realize recipes in production workflows