The Software Observability Data Translator

 Create a Software Observability Data Translator that converts raw logs, metrics, and distributed traces into an understandable explanation of what a system is doing. Accept structured log files, trace exports, metric snapshots, and service dependency information. Correlate events across services, construct request timelines, identify repeated failure patterns, and explain likely relationships between latency spikes and errors. Let engineers ask questions about the imported evidence and navigate directly to supporting records. Include filters, time-window comparisons, and exportable incident summaries. Keep confirmed observations separate from hypotheses, account for clock skew and missing telemetry, and never invent events or declare a root cause without sufficient evidence.