Staff Observability Engineer — Fleet Telemetry & Profiling

Jackalope Digital LLC; Greater London; Hybrid; GBP 325,000 - 390,000; Full time

Benefits offered by this job

Visa sponsorship available

Job summary

Anthropic is seeking Software Engineers for the Observability team to build scalable telemetry pipelines and deep visibility into multi-cluster infrastructure. You’ll own core platforms and instrument systems with OpenTelemetry and eBPF, enabling faster detection and diagnosis of issues across GPUs, TPUs, and Trainium clusters.

The role emphasizes collaboration with Research, Inference, Product, and Infrastructure teams, aiming for reliable, cost-efficient operations and AI-assisted diagnostics.

Qualifications

• Minimum hands-on experience building and operating large-scale observability or monitoring infrastructure.
• End-to-end experience with observability signals—from instrumentation through ingest to query and analysis.
• Familiarity with high-throughput telemetry pipelines and storage at scale.
• Comfort digging below the application layer into kernel, network, or hardware.
• Excellent communication and collaboration skills for cross-team visibility and incident response.
• Willingness to navigate ambiguous, high-impact technical challenges independently or with a team.

Responsibilities

• Design and build scalable telemetry ingest and storage pipelines for metrics, logs, traces, and error data across multi-cluster infra.
• Develop observability solutions for deep, low-overhead visibility into system behavior across fleets.
• Own and evolve core observability platforms, driving migrations and architectural improvements for reliability and cost efficiency.
• Build instrumentation libraries, SDKs, and eBPF-based auto-instrumentation to emit high-quality telemetry.
• Reduce MTTD/MTTR by enabling cross-signal correlation from kernel events to application traces.
• Drive fleet-wide efficiency by turning profiling data into actionable optimization insights.
• Collaborate with Research, Inference, Product, and Infrastructure teams to meet diverse needs.

Skills

Large-scale observability infra

Telemetry data pipelines

Communication skills

Debugging at kernel/network level

OpenTelemetry experience

Education

Bachelor's degree or equivalent

Tools

OpenTelemetry

eBPF

Job description

Anthropic is seeking Software Engineers for the Observability team to build scalable telemetry pipelines and deep visibility into multi-cluster infrastructure. You’ll own core platforms and instrument systems with OpenTelemetry and eBPF, enabling faster detection and diagnosis of issues across GPUs, TPUs, and Trainium clusters.

The role emphasizes collaboration with Research, Inference, Product, and Infrastructure teams, aiming for reliable, cost-efficient operations and AI-assisted diagnostics.