An AI SRE that shows its evidence — and says when it doesn't know.
Runtimez runs root-cause analysis for any workload. It gathers evidence deterministically, makes one bounded model call, and every sentence cites evidence you can click. You can ask your fleet questions from Slack, the console, the terminal or your own AI assistant.
one helm install · get, list, watch only · first findings in minutes
Root-cause analysis
It turns a pager alert into a narrative you can check.
On-demand RCA for any workload
Gathers events, logs, restarts, probes, rollouts, HPA, metrics, traces, add-on findings and mesh traffic, then makes one bounded model call. Every sentence cites its evidence.
Auto-RCA with causal chain
Root cause, then propagation, then symptom, hop by hop.
Named failure rungs
Specific causes instead of a generic "unhealthy": a liveness probe killing a container that never finished booting, scaled to zero, image pull failure, missing config reference, a PDB blocking a drain.
Hypotheses with refuting evidence
States what would disprove each hypothesis, and reports INDETERMINATE instead of making something up.
"Why is no node coming?"
For Pending replicas, names the Karpenter NodePool that refused the pod and why.
Start from the symptom
Restarting, slow, 5xx, cannot connect, memory growth, stuck rollout, not syncing. Ingress and mesh runtime evidence is scoped to the workload's own routes.
Trace bottlenecks and profiles
Critical path, self-time and per-edge latency for each trace, plus CPU and allocation profiles captured during a debug session.
Repeat incidents are free
Failure signatures are cached, so a repeat incident gets an answer instantly at no cost.
Developer productivity
Your fleet becomes a tool your team and its AI assistants can call.
Ask Runtimez
Ask the cluster a question in plain English and get a structured verdict. A scope guardrail refuses off-topic questions instead of making up an answer.
Runtimez MCP server
57 read-only tools for Claude, Cursor or any MCP client, covering inventory, changes, risk, the readiness ledger, observability and RCA. Every answer carries its syncId. Rate-limited and audited.
rtz CLI with TUI
Fleet, cluster, workload, pod and events from the terminal, plus an rtz mcp bridge.
Slack
Ask Runtimez answers and alerts arrive in the channel you already use.
AI insights feed
A ranked list of what needs attention: security, reliability, cost, hygiene, anomalies, top movers and forecasts.
How it compares
They tell you what broke. Runtimez tells you what will break, and why it broke, with citations.
One agent, the rest of the platform
Every product runs off the same read-only sweep and feeds the same ranked queue.
See it on your own cluster in under an hour.
Free for your first cluster. Read-only by default. Uninstall is one helm command.