Founding Compiler Engineer - CUDA & Rust for AI Production (SF)

Luminal; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Luminal (YC S25) is building an AI compiler to accelerate model deployment. This is a founding engineer role in downtown San Francisco, full-time on-site. You will help design the core compiler, write CUDA kernels, and drive performance improvements for production AI models.

The team values practical systems work, low-latency execution, and learning. No degree required; strong experience with GPU optimization, CUDA, and Rust is preferred.

Qualifications

• Strong proficiency in GPU model optimization.
• Proficient with CUDA for kernel development.
• Experience with Rust is highly valued.
• Egglog experience is a strong positive but not required.

Responsibilities

• Design and contribute to the core compiler architecture.
• Write CUDA kernels and optimize performance of AI models.
• Conduct model performance reviews and iterate on deployment-ready code.

Skills

GPU model optimization

CUDA

Rust

Egglog

Education

No degree required

Tools

Job description

Luminal (YC S25) is building an AI compiler to accelerate model deployment. This is a founding engineer role in downtown San Francisco, full-time on-site. You will help design the core compiler, write CUDA kernels, and drive performance improvements for production AI models.

The team values practical systems work, low-latency execution, and learning. No degree required; strong experience with GPU optimization, CUDA, and Rust is preferred.