Principal Site Reliability Engineer - AI Infrastructure Operations

Nscale is the GPU cloud engineered for AI, providing high-performance infrastructure for AI-focused companies. They are seeking a Principal Site Reliability Engineer to lead reliability strategy and operational excellence across their AI Infrastructure Operations team.

Responsibilities:

• Owning and evolving the long-term reliability strategy for Nscale’s AI and HPC infrastructure
• Designing and leading the development of large-scale control-plane systems, automation frameworks, and operational tooling
• Defining reliability standards, SLO frameworks, and operational best practices used across multiple teams
• Acting as a senior technical escalation point during critical incidents, guiding resolution and ensuring systemic fixes
• Identifying structural reliability risks and driving cross-functional initiatives to address them at the architectural level
• Partnering with Engineering, Network Operations, and Fleet Operations leadership to influence platform design and operational maturity
• Mentoring senior and mid-level engineers, raising the overall quality and effectiveness of SRE practices
• Driving measurable improvements in availability, MTTR, cost efficiency, and operational scalability

Requirements:

• 10+ years of experience in Site Reliability Engineering, Systems Engineering, or Software Engineering roles operating complex, large-scale infrastructure
• Expert-level software engineering skills, with a strong track record of building production-grade automation and systems
• Deep expertise in Linux, networking, and distributed systems design at scale
• Extensive experience debugging and resolving failures across hardware, OS, networking, and application layers
• Proven ability to lead technical initiatives across teams without direct authority
• Strong systems-thinking mindset, with the ability to balance reliability, velocity, and cost
• Deep hands-on experience with AI or HPC platforms, including GPUs, high-speed interconnects (InfiniBand/RDMA), and workload schedulers (e.g. SLURM)
• Experience designing observability systems for high-cardinality, high-throughput environments
• Familiarity with Kubernetes at scale and hybrid or bare-metal cloud architectures
• A history of driving step-change improvements in reliability, scalability, or operational efficiency