Senior AI Infrastructure Performance & Observability Engineer

CoreWeave; Bellevue (WA); On-site; USD 182,000 - 242,000; Full time

Benefits offered by this job

Health insurance

Life Insurance

Disability insurance

401(k) with employer match

Flexible PTO

Paid Parental Leave

Tuition Reimbursement

ESPP

Job summary

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to build observability systems that reveal AI infrastructure health from GPUs to distributed training. You will transform telemetry into real-time insights used by engineers, product teams, and executives.

You will own time-series data pipelines, PromQL queries, and data lake concepts (Iceberg/Parquet/Avro) to support latency, throughput, and cost metrics, collaborating with cross-functional partners.

Qualifications

• 5+ years of experience building distributed systems, observability platforms, or performance engineering tooling.
• Strong coding in Python or Go and familiarity with GPU infrastructure and performance analysis.
• Hands-on experience with Kubernetes at production scale and observability stacks (Prometheus, Grafana, OpenTelemetry).
• Fluency with time-series databases and PromQL/MetricsQL for real-time alerting and anomaly detection.

Responsibilities

• Design and build observability systems to monitor AI infrastructure health from GPUs to distributed training.
• Own time-series and metrics infrastructure to enable real-time and historical analyses.
• Build data pipelines for GPU telemetry, interconnect fabrics, and hardware health signals.
• Contribute to a data lake architecture supporting latency, throughput, and cost metrics.

Skills

Distributed systems

Observability platforms

Python

Go

Kubernetes

Prometheus

Grafana

OpenTelemetry

Perf engineering

ML infrastructure

Tools

Iceberg

Parquet

Avro

Grafana

Job description

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to build observability systems that reveal AI infrastructure health from GPUs to distributed training. You will transform telemetry into real-time insights used by engineers, product teams, and executives.

You will own time-series data pipelines, PromQL queries, and data lake concepts (Iceberg/Parquet/Avro) to support latency, throughput, and cost metrics, collaborating with cross-functional partners.