Hyground brings Kubernetes, logs, metrics, incident data, change information, and operational knowledge into one investigation workflow.

In our environment, it can query live Kubernetes data, run health checks, and preserve the evidence and reasoning for later review.

The strongest aspects are the self-hosted deployment model, the read-biased operating posture, and the ability to turn recurring SRE procedures into reusable skills.

Auto RCA is actively processing operational events, while the session history makes investigations easier to review and hand over.

Hyground also handles uncertainty reasonably well: when information is unavailable, it generally identifies the gap instead of presenting an unsupported conclusion.

Hyground reduces the manual effort needed to collect operational context during an incident a lot. This is a big plus for our 2nd level support team. Review collected by and hosted on G2.com.

Some features are still maturing. For example, file attachments did not work reliably in the version we tested, and areas such as metrics and integration diagnostics could be clearer. The team is responsive, though, and these practical issues are actively being addressed in new releases regularly. Review collected by and hosted on G2.com.