Staff+ Software Engineer, Observability
Core
Design and build scalable telemetry ingest and storage pipelines, observability solutions, and AI-assisted diagnostic tooling for Anthropic's multi-cluster GPU/TPU/Trainium infrastructure.
Role type
Staff+ Software Engineer (Observability Infrastructure)
Builds
High-throughput telemetry pipelines, fleet-wide continuous profiling, eBPF-based tracing, and agentic diagnostic tools.
Domain
Cloud Infrastructure / Observability / AI Systems
Deliverable
production ML models
Required skills
Large-scale observability infrastructure, end-to-end telemetry (instrumentation to analysis), high-throughput pipelines, kernel/network/hardware debugging, cross-signal correlation, AI-assisted diagnostics, fleet-wide efficiency optimization.
Preferred skills
eBPF-based observability, fleet-scale continuous profiling, kernel/syscall debugging, accelerator workload profiling, high-cardinality metrics systems, columnar databases (ClickHouse, BigQuery), OpenTelemetry, AI/LLMs for operational workflows.
Technologies
eBPF, OpenTelemetry, ClickHouse, BigQuery, GPU/TPU/Trainium clusters.
Responsibilities
Design scalable telemetry pipelines; build observability solutions for deep system visibility; own and evolve core observability platforms; build instrumentation libraries and SDKs; reduce MTTR via cross-signal correlation and AI tools; improve fleet efficiency via profiling insights; partner with Research, Inference, and Product teams.
Seniority
Staff+, hands-on IC with strategic impact.
