Solution Specialist, AI Runtime Services

Job summary

Responsibilities

As CoreWeave turns raw GPU capacity into production-grade AI services, it is launching new ways for customers to run, scale, and serve AI workloads, and those new services need a market-maker.

• Own the commercial and technical strategy for net new customer wins in AI runtime infrastructure, where execution performance, deployment flexibility, and operational reliability…
• Drive new business opportunities where inference latency, throughput bottlenecks, workload isolation requirements, or operational complexity are barriers to scaling AI on…
• Develop deep expertise across the AI runtime landscape (model serving architectures, execution scheduling, containerized AI workloads, and secure multi-tenant compute), using…
• Translate customer requirements around serving frameworks (e.g., vLLM, TensorRT-LLM, TGI), batching strategies, and execution isolation into specific product feedback that shapes…
• Develop deal structures, technical playbooks, and benchmark narratives that help sales and SA teams accelerate runtime-sensitive opportunities across the full spectrum of AI…

What they're looking for

• 10+ years of experience in distributed systems, ML infrastructure, or production AI engineering, with a track record of applying that expertise to drive customer outcomes and…
• 5+ years working with AI runtime systems (model serving, inference optimization, containerized workload execution, or real-time ML pipelines) in a customer-facing or deal-shaping…
• Deep working knowledge of how AI workloads execute at runtime: serving frameworks, batching strategies, GPU memory management, and the performance levers that determine…
• Experience with sandboxed and isolated execution environments (microVM architectures, container runtimes, secure multi-tenant scheduling) and how execution isolation requirements…
• Strong understanding of GPU memory hierarchies, model parallelism strategies, and how runtime architecture decisions translate into cost, latency, and scalability outcomes for…

Tech stack

• vLLM
• Triton
• Kubernetes