HPC Software Engineer for Large-Scale AI Training

Institute of Foundation Models; Abu Dhabi; On-site; AED 120,000 - 200,000; Full time

Job summary

A leading research lab in the UAE seeks a High Performance Computing Software Engineer to design and develop software for large-scale AI workloads. You will work at the intersection of HPC and machine learning, ensuring robust infrastructure for cutting-edge model training. Ideal candidates should have experience with ML frameworks and distributed communication libraries. Join a collaborative environment driving significant advancements in AI.

Qualifications

• Proven experience developing and optimizing software for large-scale ML workloads.
• Deep understanding of Linux kernel internals and accelerator kernel development.
• Proficiency with distributed communication libraries like NCCL, MPI, UCX.

Responsibilities

• Design and implement high-performance software solutions for AI/ML training.
• Optimize low-level system components including Linux kernels and GPU/accelerator kernels.
• Develop and tune communication libraries for large-scale production environments.

Skills

Software development for large-scale ML workloads

Linux kernel internals

Distributed communication libraries

Experience with ML frameworks

HPC job scheduling tools

Debugging and performance tuning

Collaborative mindset

Job description

A leading research lab in the UAE seeks a High Performance Computing Software Engineer to design and develop software for large-scale AI workloads. You will work at the intersection of HPC and machine learning, ensuring robust infrastructure for cutting-edge model training. Ideal candidates should have experience with ML frameworks and distributed communication libraries. Join a collaborative environment driving significant advancements in AI.