Description

You will engineer high-performance storage infrastructure for AI workloads, focusing on GPU IO and near speed-of-light performance.

This role is hybrid.

Responsibilities

• Develop new features and enable data storage technologies specifically for GPU IO.
• Build advanced C++ and CUDA libraries targeting extreme performance.
• Identify and eliminate bottlenecks in the IO stack, frameworks, and applications.
• Collaborate with research teams to solve complex engineering problems.

Required Skills

• 7+ years of strong coding experience in C, C++, Rust, or Python.
• Deep knowledge of Linux kernel internals, filesystems, and object storage systems.
• Proficiency with Data Structures and Algorithms.
• Experience with NVMe and related storage technologies.
• Familiarity with Cloud, Virtualization (VMware, KVM), and Container technologies.
• Solid understanding of Computer Architecture principles.
• Proficiency in Python and Bash scripting.

Preferred Skills

• Development experience in storage software (Key-Value, File systems, Object storage, Vector Databases).
• Exceptional CUDA and C++ programming skills.

Key Skills

Education

Any Graduate