ML Research Engineer (Hardware Codesign)

TL;DR

ML Research Engineer (Hardware Codesign): Shaping numerics, architecture, and technology bets for future hirify.global AI silicon in collaboration with research and hardware teams. Focus on debugging performance gaps, quantifying system architecture tradeoffs, and implementing novel numeric RTL for production.

Location: Hybrid, 3 days/week onsite in San Francisco, CA. Relocation assistance to the US is available.

Salary: $185,000–$455,000 + Equity

Company

hirify.global is an AI research and deployment company focused on developing general-purpose artificial intelligence for the benefit of humanity.

What you will do

• Build and optimize roofline simulators to deliver analyses on system architecture decisions and technology pathfinding.
• Debug performance gaps between simulation and real measurements, clearly communicating root causes and bottlenecks.
• Write emulation kernels for low-precision numerics and lossy compression schemes for model quality and efficiency tradeoffs.
• Prototype numerics modules by pushing RTL through synthesis, potentially owning RTL modules end-to-end.
• Proactively integrate new ML workloads, prototyping and driving initial evaluation of opportunities and risks.
• Understand the entire ML science to hardware optimization picture, slicing objectives into near-term deliverables.

Requirements

• Exceptional track record of high-quality technical output and a bias for shipping prototypes.
• Strong proficiency in Python, and C++ or Rust, with an intuition for correctness and extensibility.
• Experience writing Triton, CUDA, or similar, and understanding of tensor operation mapping to functional units.
• Working knowledge of PyTorch or JAX, practical understanding of floating point numerics, and ML tradeoffs of reduced precision.
• Deep understanding of transformer models, rooflines, and sharded training/inference in large-scale ML systems.
• Strong cross-functional communication skills, fostering collaborations between ML researchers and hardware engineers.

Nice to have

• Experience in large ML codebases.
• Experience writing RTL for floating point logic and understanding PPA tradeoffs.

Culture & Benefits

• Work for an AI research and deployment company dedicated to benefiting humanity.
• Opportunity to push AI system capabilities and safely deploy them to the world.
• Focus on creating AI with safety and human needs at its core.
• Commitment to equal opportunity and valuing diverse perspectives.
• Provision of reasonable accommodations for applicants with disabilities.

Hiring process

• Background checks administered in accordance with applicable law.
• Qualified applicants with arrest or conviction records will be considered.