VictoriaMetrics with NeuBird, the AI SRE: from metric to root cause
NeuBird runs MetricsQL against your VictoriaMetrics deployment at investigation time, then correlates the series that moved with the rest of your stack to find the cause.
< 5 min
Root cause, at 94% accuracy
15+
Sources queried in parallel per investigation
In place
No bulk telemetry stored
< 1 hour
To deploy on the stack you already run
Core Capabilities
From signals to solutions
Resolve
Find the series that moved first
When an alert fires, NeuBird queries VictoriaMetrics with MetricsQL for the affected service and its dependencies, breaks the change down by label to the pods, nodes or endpoints that carry it, and checks your logs, Kubernetes state and deploy history in parallel. It returns the causal chain with the queries and series attached.
- Narrows a spike to the label values that carry it
- Orders the series that moved, so the first change stands out
- Root cause in under 5 minutes at 94% accuracy, posted where your team works
Prevent
Catch saturation while it builds
NeuBird watches VictoriaMetrics trends against each service’s dependencies and flags saturation and capacity bottlenecks 30 to 60 minutes before an alert would fire.
- Flags CPU, memory, disk and queue depth trending toward limits
- Surfaces services that expose no latency or error metrics
- Recommends missing SLOs for the services you already measure
Operate
Less noise between incidents
NeuBird reviews the metrics behind recurring alerts and recommends tuning for the rules and thresholds that page people for nothing.
- Identifies alert rules whose thresholds no longer match normal load
- Spots recurring patterns across past incidents
- Remembers every verified fix, with no bulk telemetry stored
Better Together
VictoriaMetrics + NeuBird
| Capability | VictoriaMetrics | NeuBird |
|---|---|---|
| High-volume, Prometheus-compatible metrics storage | ✓ | Reads it in place |
| MetricsQL queries across series and labels | ✓ | Runs them during each investigation |
| Correlation with logs, Kubernetes and code changes | Through other tools | 15+ sources in parallel |
| Root cause with cited evidence | Query results | Under 5 minutes, 94% accuracy |
| Staged fix or rollback | Not in scope | Behind configurable human approval gates |
| Memory of past fixes | Not in scope | Verified fixes, no bulk telemetry stored |
Ecosystem
Works across your entire stack
VictoriaMetrics is one piece of the picture. NeuBird queries it in place and correlates it with every other connected tool, and every agent in the center inherits the connection.
Observability
- VictoriaLogs
- VictoriaTraces
- Grafana
- Prometheus
Infrastructure
- Kubernetes
- AWS
- Google Cloud
- Microsoft Azure
Incident management
- PagerDuty
- Opsgenie
- incident.io
Code and change
- GitHub
- GitLab
- Argo Workflows
FAQ
Common questions
How does NeuBird connect to VictoriaMetrics?
You create a dedicated vmauth user for NeuBird and enter its credentials as an HTTP Basic auth token, with the vmauth URL for metrics including its route prefix. A deployment inside your network can be published as a Teleport application instead. NeuBird queries VictoriaMetrics at investigation time and stores no bulk telemetry.
What VictoriaMetrics data does NeuBird use?
The metric time series and labels that the vmauth user can read, queried with MetricsQL. VictoriaLogs and VictoriaTraces are separate connections.
Does NeuBird change anything in VictoriaMetrics?
NeuBird uses VictoriaMetrics as an investigation source. Any fix it proposes is staged for approval through the Suggest, Recommend, Act policy dial, set per action, with configurable human approval gates.
Is NeuBird a replacement for VictoriaMetrics?
No. VictoriaMetrics keeps storing and serving your metrics. NeuBird is the AI SRE that investigates what those metrics show, alongside every other tool you connect.
Get Started
Connect VictoriaMetrics to the center.
VictoriaMetrics gives you the data. Connect it to NeuBird once, and every agent that touches production inherits it, including the ones you build yourself.
