Node.js OpenTelemetry Integration Expert
Budget: $2 – $8 USD
I need a senior-level Node.js engineer to weave OpenTelemetry into my existing service so I can watch real-time performance trends, isolate slow calls, and correlate everything down to individual log lines. The brief is full-spectrum instrumentation—tracing, metrics, and logs—using the official OpenTelemetry SDK for Node.js.
You’ll be expected to:
• Add auto and manual instrumentation where the auto hooks fall short, with clean context propagation across every async boundary.
• Need to use library like winston/bunyan for it with zero auto instrumentation
• Configure the collector/exporter pipeline; the backend hasn’t been chosen yet, so I’m open to your recommendation as long as it accepts OTLP.
• Keep overhead low and prove there’s no noticeable latency hit in production-like benchmarks.
• Deliver a concise README that shows how to spin everything up locally, view traces and metrics in the selected backend, and map logs back to a given trace ID.
Acceptance criteria: the dashboard (or viewer you propose) must show end-to-end traces for every HTTP request, key service-level metrics (latency, throughput, error rate), and logs linked to each span. All work will be merged via PR, fully reviewed and documented.
You’ll be expected to:
• Add auto and manual instrumentation where the auto hooks fall short, with clean context propagation across every async boundary.
• Need to use library like winston/bunyan for it with zero auto instrumentation
• Configure the collector/exporter pipeline; the backend hasn’t been chosen yet, so I’m open to your recommendation as long as it accepts OTLP.
• Keep overhead low and prove there’s no noticeable latency hit in production-like benchmarks.
• Deliver a concise README that shows how to spin everything up locally, view traces and metrics in the selected backend, and map logs back to a given trace ID.
Acceptance criteria: the dashboard (or viewer you propose) must show end-to-end traces for every HTTP request, key service-level metrics (latency, throughput, error rate), and logs linked to each span. All work will be merged via PR, fully reviewed and documented.