Site Reliability Engineer

The role

The Staff Site Reliability Engineer will design and operate reliable managed AI services. This role focuses on serving and scaling LLM workloads and building automation for distributed AI pipelines.

What you will do

• Design and operate reliable managed AI services with a focus on serving and scaling LLM workloads
• Build automation and reliability tooling to support distributed AI pipelines and inference services
• Define, measure, and improve SLIs/SLOs across AI workloads
• Collaborate with AI, platform, and infrastructure teams to optimize large-scale training and inference clusters
• Automate observability by building telemetry and performance tuning strategies for latency-sensitive AI services
• Investigate and resolve reliability issues in distributed AI systems using telemetry, logs, and profiling
• Contribute to the architecture of next-generation distributed systems purpose-built for AI-first environments

Required qualifications

• Strong software engineering background with experience building production-grade systems
• Demonstrated experience in distributed systems design and implementation
• Hands-on work with large language models (LLMs) or AI/ML infrastructure
• SRE mindset and experience including defining and measuring SLIs/SLOs
• Experience building monitoring and observability systems
• Experience driving performance and reliability improvements
• Experience designing fault-tolerant systems and automated testing strategies
• Proficiency in at least one modern programming language (Python, Go, Java, C++)
• Deep working knowledge of Kubernetes (cluster architecture, networking, scheduling, and workload management)
• Strong collaboration and communication skills
• Ability to thrive in a fast-paced, mission-driven environment

Nice to have

• Experience scaling inference or training workloads for LLMs

Benefits

• Pension contributions
• Additional perks to support work-life balance

About the company

Crusoe is a vertically integrated AI infrastructure company that owns and operates each layer of the stack to power AI workloads. The company uses an energy-first approach to make AI infrastructure better and faster.