axiom-sre

Hypothesis-driven incident response requires engineers to state, test, and disprove hypotheses using data queries and evidence. This systematic triage approach enforces data-backed investigations for root-cause analysis, stabilizing systems while logging findings to memory for future learning.

You can execute Axiom APL queries for live debugging by first running initialization scripts, then using scripted discovery workflows to query logs. This generates shareable links to artifacts, enabling rapid collaboration and evidence gathering during incident triage.

Yes, you need jq, git, and curl installed. These dependencies support the scripted discovery workflows, evidence recording, and memory KB integration required to enforce initialization and execute data-driven incident investigations properly.

The best way to prepare a postmortem is using a persistent memory knowledge base that logs findings, patterns, queries, and runbooks during the incident. This accelerates postmortem preparation by capturing evidence and disproved hypotheses throughout the investigation.

Yes, you can apply USE and RED methodologies for systematic triage alongside Axiom log integration. The workflow prioritizes fixes by evaluating golden signals and executing APL queries to gather evidence, ensuring hypotheses are tested against actual system metrics.

Initialization is required before querying logs to enforce discovery steps and ensure access to configured deployments and memory. This structured approach prevents unstructured troubleshooting, requiring data-driven, hypothesis-driven investigations that record evidence for future learning.