Design and implement distributed tracing systems using OpenTelemetry, Jaeger, Zipkin, and Tempo to track requests across microservices and pinpoint latency bottlenecks.
Understanding what happens to a request as it travels through dozens of microservices is one of the hardest problems in modern cloud infrastructure. The Distributed Tracing Engineer helps platform engineers, SREs, and backend developers instrument their systems, design trace collection pipelines, and extract actionable insights from distributed trace data.
This assistant guides you through the full distributed tracing stack: selecting and configuring instrumentation libraries such as OpenTelemetry SDKs for your language and framework, setting up trace collectors and exporters, choosing and deploying a tracing backend like Jaeger, Zipkin, Grafana Tempo, or AWS X-Ray, and configuring sampling strategies that balance observability coverage with data volume and cost.
When you describe a latency problem, a mysterious error spike, or a slow user-facing transaction, the assistant helps you interpret trace waterfalls, identify span anomalies, and correlate trace data with logs and metrics to build a complete picture of what went wrong and where. It explains how to read flame graphs, how to diagnose cascading failures across service boundaries, and how to attribute latency to specific downstream calls.
The assistant also helps you design trace instrumentation that goes beyond automatic framework-level spans. It generates custom span code, baggage propagation patterns, and context injection logic for messaging systems like Kafka or RabbitMQ, where trace context does not propagate automatically. It advises on head-based versus tail-based sampling and helps you write sampling rules that capture the traces you care about most.
Ideal users include platform engineers rolling out observability stacks for microservices architectures, SREs investigating production incidents where root cause spans multiple services, backend engineers adding tracing to an existing codebase for the first time, and engineering teams migrating from proprietary APM tools to open standards. The Distributed Tracing Engineer helps you stop guessing where time goes in your system and start knowing.
Sign in with Google to access expert-crafted prompts. New users get 10 free credits.
Sign in to unlock