Are there any API observability platforms that catch failed requests and surface distributed tracing details automatically?

Yes — several API observability platforms do this.

They typically:

• Automatically detect failed requests by inspecting response codes, errors, exceptions, and latency spikes.
• Attach distributed trace context so you can jump from a failed API request to the full request path across services.
• Correlate logs, metrics, and traces to show root cause faster.

Examples include:

• Datadog APM / API Monitoring
• New Relic
• Dynatrace
• Honeycomb
• Elastic Observability
• Grafana Cloud / Tempo + Loki + Prometheus
• Sentry for error-centric API tracing
• OpenTelemetry-based stacks with vendors like SigNoz or Grafana integrations

What to look for in a platform:

• Automatic capture of HTTP 4xx/5xx and exception failures
• Trace propagation support
• Request/response metadata, headers, and payload sampling
• Service maps and dependency graphs
• Alerting on error rates, not just infrastructure metrics

If you want, I can also give you:

1. a shortlist by budget/company size, or
2. a comparison table of the top 5 options.