Senior Systems Performance Engineer

What you'll do

• Develop and implement complex automated test plans for GPU accelerated computing products.
• Perform system architecture design, performance modeling, and estimation across new models and packages.
• Execute GPU SKU bring up, validation, and model enablement processes.
• Create system-level stress and performance testing strategies using Deep Learning and AI applications.
• Develop and run real-world ML and LLM workloads for product validation.
• Debug complex hardware and software systems within a physical laboratory environment.
• Utilize tools like Dynamo, TensorRT, Slurm, and BCM to validate system performance.
• Write Python code to run simulators and automate testing workflows.

What we're looking for

• Ability to work on site in a hardware lab environment 5 days a week.
• BSEE, BSCE, or equivalent experience.
• 5+ years of experience in validating and debugging complex systems.
• Experience developing and running real-world ML/LLM workloads.
• Mandatory skills in Dynamo, TensorRT, Slurm, and BCM.
• Proficiency in Cuda, Cublas, Cutlass, and Python programming.
• Deep understanding of computing architectures and experience with datacenter products including networking and storage.
• Preferred knowledge of vLLM and SG Lang.