Edge Transformer Inference Tech Lead

OpenAI; San Francisco (CA); Hybrid; USD 130,000 - 180,000; Full time

Job summary

A leading AI research firm in San Francisco is seeking a Technical Lead to join its Future of Computing Research team. This role involves evaluating silicon platforms and optimizing model architectures while working in a hybrid model. Ideal candidates have expertise in evaluating workloads on accelerators, understanding transformer models, and leading teams focused on performance-critical software. The position offers relocation assistance and is centered on deploying cutting-edge AI technology responsibly and effectively.

Qualifications

• Experience with GPUs, NPUs, or other specialized accelerators.
• Understanding of memory bandwidth requirements in transformer models.
• Experience in optimization of inference engines and hardware-aware ML pipelines.

Responsibilities

• Evaluate silicon platforms for on-device deployment.
• Co-design model architectures with research teams.
• Analyze system performance, optimizing trade-offs.
• Lead a team responsible for low-level inference stack.
• Transform research capabilities into practical applications.

Skills

Experience evaluating or deploying workloads on GPUs

Understanding transformer model performance characteristics

Designing high-performance compute systems

Building or leading teams on performance-critical software

Experience with CUDA kernels

Job description

A leading AI research firm in San Francisco is seeking a Technical Lead to join its Future of Computing Research team. This role involves evaluating silicon platforms and optimizing model architectures while working in a hybrid model. Ideal candidates have expertise in evaluating workloads on accelerators, understanding transformer models, and leading teams focused on performance-critical software. The position offers relocation assistance and is centered on deploying cutting-edge AI technology responsibly and effectively.