Software Engineer, Kernel Programming Model

Job Title

Software Engineer, Kernel Programming Model

Role Summary

Design and implement the runtime and integration layers that allow custom compute kernels to run as native extensions within the PyTorch ecosystem. Work on architectural abstractions (virtual ISA) to expose low-level hardware control while preserving tensor-level programmability. Build tooling, specifications, and reference implementations to support kernel developers and an external developer ecosystem.

Experience Level

Mid-level. No explicit years of experience provided.

Responsibilities

Main responsibilities include:

• Design and implement PyTorch-native integration layers and runtime support for custom kernels.
• Define and implement a virtual ISA and architectural abstractions to optimize performance while maintaining programmability.
• Develop core programming tools, reference implementations, and technical specifications to enable kernel development and community contributions.

Requirements

Must-have technical skills and experience; preferred items listed separately.

• Systems programming experience in Rust, C++, or Go with strong low-level programming skills.
• Deep understanding of computer architecture, including ISA, SIMD, and memory hierarchies.
• Experience designing and implementing clean, robust programming interfaces and runtime components.
• Preferred: experience with low-latency asynchronous execution models, kernel-level performance optimizations, hardware-software co-design, compiler infrastructure (LLVM/MLIR), or language/DSL design.
• Preferred: history of engaging developer communities through open-source contributions or technical evangelism.

Education Requirements

Bachelor's degree in Computer Science or a related technical field, or equivalent practical experience. Master's or PhD in Computer Science or a related technical field preferred; equivalent practical experience is also acceptable.

About the Company

Company: FuriosaAI

Headquarters: Seoul, South Korea

FuriosaAI develops high-performance, energy-efficient AI inference hardware and software. Founded in 2017 by semiconductor and AI engineers, the company builds AI-native compute platforms to reduce AI energy and operational costs and operates globally with offices in Korea, Silicon Valley, and an R&D lab in Lisbon.