Magic’s mission is to build safe AGI that accelerates humanity’s progress on the world’s most important problems. We believe the most promising path to safe AGI lies in automating research and code generation to improve models and solve alignment more reliably than humans can alone. Our approach combines frontier-scale pre-training, domain-specific RL, ultra-long context, and inference-time compute to achieve this goal.

About the role:

As a Software Engineer on our Supercomputing Platform & Infrastructure team, you will design and build resilient and optimized solutions for AI workloads on massive Computing Clusters.

What you might work on:

Work closely with the training and inference teams to deliver high performance and reliability across storage, networking, and distributed computing designs.
Build the software stack to run massive-scale (thousands of GPUs), highly available supercomputing infrastructure
Troubleshoot and resolve complex issues across hardware accelerated devices, networking, storage subsystems (local NVMe/Block Storage/NFS), OS, drivers and cloud environments, and automate detection and recovery processes
Operate data-intensive workloads at petabyte-scale
Increase the ease-of-use and self-serviceability of the compute platforms at Magic through top-notch documentation and developer workflow design
Investigate and resolve incidents across security and availability

What we’re looking for:

Experience working with production GPU deployments, data-intensive applications, large-scale model training and HPC
Strong understanding of networking-, storage- and data-related technologies
Experience with GCP, AWS, Azure, OCI or similar cloud platforms
Strong software engineering skills
Strong IaC knowledge with extensive experience in Terraform, Pulumi, AWS CDK/CloudFormation or similar

Magic strives to be the place where high-potential individuals can do their best work. We value quick learning and grit just as much as skill and experience.

Our culture: