Staff Software Engineer, Runtime Systems

Role details

CoreWeave is The Essential Cloud for AI™. Built for pioneers by pioneers, CoreWeave delivers a platform of technology, tools, and teams that enables innovators to build and scale AI with confidence. Trusted by leading AI labs, startups, and global enterprises, CoreWeave combines superior infrastructure performance with deep technical expertise to accelerate breakthroughs and turn compute into capability. Founded in 2017, CoreWeave became a publicly traded company (Nasdaq: CRWV) in March 2025. Learn more at www.coreweave.com . What You’ll Do: The Physical AI Engineering Team at CoreWeave is building the software and infrastructure that enables demanding AI, simulation, robotics, and engineering workloads to run reliably at scale. As these workloads become more complex, the challenge is no longer simply providing compute. We need to make heterogeneous workloads easier to execute, observe, reproduce, and move across different systems without hiding the capabilities or semantics of the infrastructure underneath them. About the role: We’re seeking a Staff Software Engineer, Runtime Systems to help design and build this layer. This is a hands-on systems engineering role at the intersection of distributed systems, runtimes, workflow execution, programming language concepts, and large-scale compute infrastructure . You’ll work on the foundations that allow complex workloads to move from an abstract description into reliable execution across systems such as Kubernetes, Argo, OSMO, and future execution environments. A major part of the role is deciding where abstraction is useful — and where it creates more complexity. Rather than building another universal workflow engine, you’ll help establish clear contracts between our platform and the systems that execute work, while preserving native capabilities. You’ll operate across architecture and implementation: defining contracts, writing production software, validating assumptions against real workloads, and working closely with platform and infrastructure teams. In this role, you will: Runtime Systems & Architecture Design and build runtime components for complex AI, simulation, and engineering workloads. Define abstractions for workloads, execution environments, dependencies, state, capabilities, and failure. Design interfaces between higher-level services and systems such as Kubernetes, Argo, and OSMO. Establish clear boundaries around which systems own state, decisions, and side effects. Make architectural decisions balancing simplicity, extensibility, performance, and operational reality. Execution Models & Contracts Define durable contracts between workload definitions, control-plane services, and execution backends. Develop typed representations and schemas that allow workloads to be transformed safely across systems. Design compatibility and evolution mechanisms for those contracts. Build conformance and validation mechanisms that make guarantees executable rather than dependent on documentation. Reason deeply about retries, partial failure, idempotency, cancellation, dependencies, and uncertain outcomes. Hands-On Systems Engineering Write production-quality software for critical runtime and control-plane components. Build adapters and integrations for heterogeneous execution environments. Diagnose behaviour across application, orchestration, cluster, and infrastructure boundaries. Improve the reliability, observability, and debuggability of distributed workload execution. Work closely with Go, Kubernetes, and infrastructure engineers to turn architecture into production systems. Performance & Experimentation Develop rigorous ways to understand workload performance across large-scale GPU infrastructure. Design experiments that separate real performance gains from noise, warm-up effects, scheduling behaviour, and stragglers. Build repeatable workload and benchmark environments. Use evidence from real execution to challenge assumptions and guide platform development. Technical Leadership Lead ambiguous systems problems where the correct architecture is not yet known. Reduce complex problems into smaller contracts and mechanisms that can actually be implemented. Challenge unnecessary abstraction and simplify designs where complexity has outgrown its value. Influence technical direction across teams without requiring direct authority. Mentor engineers and contribute to technical hiring and engineering standards. Who You Are: Significant experience building complex systems software, distributed infrastructure, runtimes, workflow systems, or adjacent technology. Deep expertise in at least one of: Distributed systems Runtime systems Workflow or execution engines Programming languages, compilers, or interpreters Cluster scheduling and orchestration High-performance or systems software Strong software engineering fundamentals and production coding ability. Experience designing APIs, protocols, schemas, or contracts between independently evolving systems. Strong understanding of distributed-system failure modes, state, authority, retries, concurrency, and side effects. Strong technical judgement around when abstraction helps and when it simply moves complexity elsewhere. Comfortable entering unfamiliar technical domains and building depth quickly. Strong communication skills and experience influencing architectural decisions across teams. Experience with some of the following would be valuable, but is not required: Go, Rust, C/C++, or Python. Kubernetes and containerised infrastructure. Argo, OSMO, Temporal, Ray, Kubeflow, or similar systems. GPU clusters or large-scale AI infrastructure. High-performance computing. Simulation, robotics, autonomous systems, or Physical AI. Programming language or compiler research. Performance engineering. Cloud infrastructure at scale. Platforms designed to be operated by autonomous software or AI agents. Wondering if you’re a good fit? We believe in investing in our people, and value candidates who can bring their ow