Description

You will design and build high-performance parallel processing software, defining how data processing engines leverage GPU and SIMD-capable CPU parallelism for large-scale data.

This role is on-site.

Responsibilities

• Design performance-enhancing improvements for acceleration-enabled operators and define GPU/SIMD utilization strategies.
• Develop, test, optimize, and maintain parallel data operator implementations across various acceleration hardware.
• Drive best practices in code reviews, testing, CI/CD, debugging, and performance optimization.
• Integrate parallel software components with execution engine teams to ensure seamless functionality.

Required Skills

• Strong programming expertise in C++ and CUDA/GPGPU computing.
• Hands-on experience with parallel computing on GPUs and CPUs.
• Solid understanding of CPU architecture, SIMD, and memory hierarchy.
• Experience in performance optimization, profiling, and debugging.
• Familiarity with CI/CD pipelines and modern software development practices.
• Knowledge of Spark/distributed computing frameworks.
• 5+ years of experience in software engineering.

Key Skills

Education

Any Graduate