Senior AI Infra Performance & Observability Engineer

CoreWeave; Sunnyvale (CA); On-site; USD 182,000 - 242,000; Full time

Benefits offered by this job

Medical, dental, vision insurance

401(k) with employer match

Paid Parental Leave

Catered lunch in office

Job summary

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to build observability systems for AI infrastructure. You will own detection, diagnosis, and surface signals across data centers, turning telemetry into actionable insights for engineers and executives.

You will design time-series infrastructure, validate GPU fabric telemetry, and help shape a data lake architecture to support latency distributions and cost-per-token metrics.

Qualifications

• 5+ years designing distributed systems or observability platforms for infrastructure or ML workloads.
• Strong coding in Python or Go; C++ is a plus.
• Hands-on Kubernetes at production scale with CI/CD and observability stacks.
• Fluency in time-series queries (PromQL/MetricsQL) for real-time alerting and anomaly detection.
• Familiarity with data lake architectures and table formats (Iceberg, Parquet, Avro).
• Comfort working close to hardware and workload behavior in GPU infrastructure.

Responsibilities

• Design and build systems that continuously assess AI infrastructure health and performance.
• Own TSDB layer and write/optimize PromQL/MetricsQL for alerting and trend analysis.
• Develop pipelines for GPU telemetry, interconnect fabric, and throughput signals.
• Design a performance data lake with Iceberg, Parquet, and Avro support.
• Profile and tune query engines for low latency and high freshness SLAs.
• Build self-service BI views as needed, downstream of the core insight work.

Skills

Distributed systems

Observability

Performance tooling

Python

Go

Kubernetes

Prometheus

Grafana

OpenTelemetry

PromQL

Tools

Iceberg

Parquet

Avro

Job description

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to build observability systems for AI infrastructure. You will own detection, diagnosis, and surface signals across data centers, turning telemetry into actionable insights for engineers and executives.

You will design time-series infrastructure, validate GPU fabric telemetry, and help shape a data lake architecture to support latency distributions and cost-per-token metrics.