Hands-on Leadership Role - Inference Engine Optimization

About

Kog is hiring for a hands-on leadership role focused on optimizing its inference engine at the hardware boundary.

Responsibilities

• Optimize the inference engine at the hardware boundary.
• Develop GPU kernels.
• Understand and leverage memory hierarchy.
• Influence architectural decisions.
• Collaborate closely with a small team on critical aspects determining token generation speed.
• Guarantee measurable performance improvements.

Qualifications

• Expertise in GPU kernel development.
• Understanding of memory hierarchy.

Benefits

• Remote-first working conditions.
• Competitive compensation aligned with the top Paris AI market standards.