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 an engineer on the Supercomputing Platform & Infrastructure team, you will design, build, and operate the large-scale GPU infrastructure that powers Magic’s model training and inference workloads.

A core part of this role is building and maintaining our infrastructure using Terraform-driven infrastructure-as-code practices, ensuring reproducibility, reliability, and operational clarity across clusters spanning thousands of GPUs.

Magic’s long-context models create sustained pressure on compute, networking, and storage systems. Long-running distributed jobs, high-throughput data movement, and strict availability requirements demand infrastructure that is automated, observable, and resilient by design. You will own the systems and IaC foundations that make this possible, including the Kubernetes (K8s) environments that coordinate workloads across our GPU infrastructure.

This role can evolve into broader ownership of supercomputing platform architecture, shaping how Magic scales GPU clusters and infrastructure reliability as model workloads grow.

What you’ll work on

• Design and operate large-scale GPU clusters for training and inference

• Build and maintain infrastructure using Terraform across cloud and hybrid environments

• Deploy, operate, and optimize K8s clusters used to schedule and manage AI workloads

• Develop modular, scalable IaC patterns for compute, networking, and storage provisioning

• Improve deployment reproducibility, environment consistency, and operational safety

• Optimize networking and storage systems for high-throughput AI workloads

• Automate fault detection and recovery across distributed clusters

• Debug complex cross-layer issues spanning hardware, drivers, networking, storage, OS, and cloud

• Improve observability, monitoring, and reliability of core platform systems

What we’re looking for

• Strong software engineering skills with experience building production infra systems

• Deep, hands-on experience with Terraform, including module design, state management, environment isolation, and large-scale deployments

• Experience operating production GPU infrastructure or high-performance distributed systems

• Strong understanding of networking and storage systems

• Experience with major cloud platforms (GCP, AWS, Azure, OCI, etc.)

• Track record of owning production-critical infrastructure end-to-end

Compensation, benefits, and perks (US):

• Annual salary range between $200K - $550K depending on experience

• Equity is a significant part of total compensation, in addition to salary

• 401(k) plan with 6% salary matching

• Generous health, dental and vision insurance for you and your dependents

• Unlimited paid time off

• Visa sponsorship and relocation stipend to bring you to SF, if possible

• A small, fast-paced, highly focused team

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.