Principal Software Engineer, Compute

About Radiant

Radiant is an El Segundo, CA-based startup building the world’s first mass‑produced, portable nuclear microreactors. The company’s first reactor, Kaleidos, is a 1‑megawatt, fail‑safe microreactor that can be transported anywhere power is needed and run for up to 5 years without refueling. Portable nuclear power with rapid‑deploy capability can replace similar‑sized diesel generators and provide critical asset support for hospitals, data centers, remote sites, and military bases.

Radiant’s unique, practical approach to nuclear development leverages modern software engineering to rapidly deliver safe, factory‑built microreactors that use existing, well‑qualified materials. Founded in 2020, Radiant is on track to test its first reactor at the Idaho National Laboratory this summer, with initial customer deliveries beginning in 2028.

Radiant is seeking a driven Principal Software Engineer to own on‑site High‑Performance Compute infrastructure, scientific computing environments, and automation projects in tight collaboration with hard‑science users and cloud infrastructure engineers. You will work closely with the software team to design scalable, secure, and resilient compute practices, tools, and systems across the entire organization.

As a technical lead, you will define team scope, shape individual responsibilities, and serve as the primary liaison between engineering and software teams, acting as the subject‑matter expert on HPC workflows, simulation tooling, and compute optimization. The ideal candidate is patient, organized, and comfortable managing high volumes of cross‑disciplinary requests, capable of diving deep into complex legacy stacks and synthesizing findings.

The infrastructure you manage, the pipelines you build, and the developer‑productivity culture you establish will help you design, run, and mass‑produce the first high‑temperature gas‑cooled portable microreactor ever to be commercialized.

• Lead HPC initiatives as driven by the software org, establishing responsibilities, project scope, and technical mentorship.
• Own workflows, tooling, and performance for scientific computing, including Ansys, STAR‑CCM+, and Abaqus, covering licensing, environment setup, job orchestration, and results infrastructure.
• Dive deep into HPC software, workload scheduling, data movement, storage hierarchies, and compute environments to build robust, high‑throughput Linux systems.
• Serve as the go‑between for engineering teams and software, fielding HPC questions, simulation software issues, and infrastructure needs from nuclear, thermal, materials, mechanical, and electrical engineers, translating them into actionable work.
• Triage inbound infrastructure requests, HPC/MPI/Linux debugging, job failure analysis, and shell and systems mentorship, while prioritizing effectively and communicating clearly across stakeholders.
• Modernize legacy scientific computing systems and tooling, migrating to current stacks to improve maintainability, performance, and developer experience.
• Design and maintain networking infrastructure for distributed simulation systems, optimizing data transfer between HPC clusters.
• Partner with Dev Ops engineers to architect and maintain systems across AWS and on‑premises Linux environments, ensuring high availability, security, and performance for mission‑critical workloads.
• Architect tools supporting build systems, testing frameworks, deployment automation, and developer environments.

• Bachelor’s degree in Computer Science, Engineering, Physics or a related technical field.
• 8+ years of professional experience in HPC, scientific computing, infrastructure, or platform engineering.
• Deep Linux and sysadmin fluency, file systems, process management, and networking, with a hard‑science approach to problem‑solving.
• Hands‑on experience with HPC environments, simulation workloads, or academic/research computing (SLURM, PBS, MPI, Lustre, EFS, AWS Parallel Cluster, or similar).
• Strong proficiency in one or more languages:
  Python, Golang, Rust, C#, or C/C++.
• Strong code review skills, including the ability to read stack traces…