Description

You will build performance models to prototype Generative AI algorithms and drive model-hardware co-design.

Responsibilities

• Develop high-fidelity performance models to prototype emerging algorithmic techniques in Generative AI.
• Design targeted optimizations for inference deployment to maximize accuracy, throughput, and interactivity.
• Model end-to-end performance impact of GenAI workflows, including Agentic Pipelines and inference-time compute scaling.
• Quantify performance benefits of optimizations to guide software and hardware roadmaps.
• Collaborate with DL researchers, hardware architects, and software engineers.

Required Skills

• Master's degree in Computer Science, Electrical Engineering, or a related field.
• 3+ years of experience in system evaluation of AI/ML workloads or performance analysis and modeling.
• Strong background in computer architecture, roofline modeling, queuing theory, and statistical performance analysis.
• Solid understanding of LLM internals, including attention mechanisms, FFN structures, model parallelism, and inference serving.
• Proficiency in Python for simulator design and data analysis.
• Experience defining metrics, designing experiments, and visualizing large performance datasets.
• Ability to distill complex analyses into clear recommendations for technical and non-technical stakeholders.

Preferred Skills

• Proficiency in C++.
• Experience with GPU computing (CUDA).
• Experience with deep learning frameworks including PyTorch, TRT-LLM, VLLM, and SGLang.

Key Skills

Education

Master's degree