High Performance Computing Software Engineer - Supercomputing

About the role

IFM is building the foundational compute infrastructure for large-scale AI and computational science, and is looking for a High Performance Computing Software Engineer to design, develop and operate the software systems that run its workloads. The role sits at the intersection of high-performance computing and machine learning, supporting training of models across 1000+ GPUs.

Key responsibilities

• Design and implement high-performance, distributed software for large-scale AI/ML training.
• Optimise low-level system components such as the Linux kernel, GPU/accelerator kernels and interconnects.
• Develop and tune communication libraries such as NCCL, MPI, UCX, RCCL and RDMA-based systems.
• Partner with ML researchers and engineers to support frameworks like PyTorch, MegatronLM and DeepSpeed in production environments.
• Contribute to scheduling, orchestration and job management systems, including Slurm and Kubernetes.
• Debug and resolve issues across the stack, from kernel to container to model.
• Work with hardware vendors, upstream open-source communities and internal teams to improve performance and reliability.

Required profile

• Proven experience developing and optimising software for large-scale ML workloads (1000+ GPUs preferred).
• Strong knowledge of HPC job scheduling and orchestration tools.
• Excellent debugging and systems performance tuning skills.
• Collaborative mindset focused on shared success and technical excellence.

Required skills

• Linux kernel internals and accelerator (GPU) kernel development.
• Distributed communication libraries: NCCL, RCCL, MPI, UCX, SHARP, Libfabric.
• ML frameworks such as PyTorch, TensorFlow, JAX or MegatronLM.