Why Epok
One evidence engine.
Your team and your agents.
Epok ingests your telemetry and turns it into evidence you can check. Recognize incidents, investigate root causes, assess recovery and forecast trouble. Start with the worked scenarios in the live demo ↗, then test the fit with your own telemetry.
From owned telemetry to checkable action.
During an incident
- Recommends the reversible fix
- Recovery starts with the change and its blast radius while root-cause analysis continues. For known failure shapes, runbooks and built-in playbooks supply reversible actions with guards, such as when a rollback won't recover the service. We recommend; you execute.
- Shows when the evidence is incomplete
- Epok commits a cause only when supporting evidence is present. Otherwise it abstains, explains what is missing and says what to check next. Coverage distinguishes a finding, none in retained data, and a check that did not run.
- One investigation surface
- Logs, metrics, traces, infrastructure, RUM and replay meet on one canvas: RCA, triggering charts, notification status, cascade timeline, causal graph, deep RCA, postmortem and architecture context. Inspect the evidence and its coverage across the incident.
- Follow telemetry as it arrives
- Stream live logs and trace spans and search the last hour without waiting on a query.
- Ask in English
- Ask a question in plain English, then inspect the generated query and source results alongside the answer.
Every day
- Alert lifecycle + ack
- Operators acknowledge with comment + timeout; escalation pauses on ack and re-fires when the timeout expires. Unacked criticals re-page on a configurable cadence. Bulk-ack for burst incidents. Activity log per action.
- Alert-noise reduction
- Similar errors are collapsed into one alert using semantic similarity. Patterns that re-appear after going quiet don't re-page; they update the existing alert.
- Owned ingest with open shippers
- Epok owns the pipeline from raw telemetry to evidence. Use open shippers — no proprietary Epok agent for server-side telemetry. Follow the integration guide for the protocol and fields your workload needs.
- Evidence for your team and agents
- Evidence items carry stable IDs, typed values, receipts and coverage, including what was not seen. Your team follows them in the app, Slack and PagerDuty. Agents query the same evidence through the MCP gateway, with access on request.
Before and buying
- Forecasts the breach — before it pages
- For saturation, exhaustion and capacity trends, Epok projects a breach with an ETA and confidence. Follow the supporting evidence and its data window, coverage and uncertainty to assess preventive action.
- Predictable pricing
- Team starts at $199/month with 1 TB included. Growth includes 4 TB for $599/month. Published overage is $0.20/GB; daily ingest limits apply. No separate host, query or cardinality charge.
Explore an incident. No signup.
Inspect the demo, then connect a representative service in the 14-day trial. See pricing.