Capabilities · Anomaly Detection
Transaction Anomaly Detection
HealthyZero-shot detectionEndpoint ledger-anomaliesProduction v2Challenger v3
Production constraints
Constraints drive candidate selection and deployment decisions.
P95 latency
<
ms
Alerts / day
<
Alert precision
>
%
Cost / 1K
<
$
Challenger v3: satisfies all constraints
Candidate eligibility preview
✓Managed AAlert precision 84% · 14ms · $0.03Selected
✕Managed BAlert precision 86% · 48ms · $0.19cost $0.19
✓Managed CAlert precision 62% · 2ms · $0.01Eligible
Capability
- Endpoint
- ledger-anomalies
- Task
- Unsupervised anomaly detection
- Objective
- Detect unusual activity in led…
- Mode
- Zero-shot detection
- Backend
- Managed by Zero Grad
- Examples
- ledger_events.parquet
- Available history
- 22.6M observations
- In-context examples
- 8-week baseline
- Target
- — (unsupervised)
- Production version
- v2
By default you manage capabilities and versions, not models. Turn this on to see which engine family backs each candidate.
Optimization mode
Preview
Optimize for
Allowed actions
When a better model appears, Autopilot evaluates it against your data, shadows it, and rolls it out within your constraints — your application changes zero code.
Where should this capability run?
Enterprise controls
Data residency
Pin data and inference to a region (US, EU)
Private networking
PrivateLink / Private Service Connect endpoints
Customer-managed encryption keys
Bring your own KMS keys
Audit logs
Every deployment, promotion and config change, exportable
SSO / SAML
Okta, Google Workspace, Azure AD
RBAC
Separate who can deploy, promote and change guardrails
Dedicated endpoints
Isolated capacity with reserved throughput
Custom retention
Configure how long inputs and predictions are kept
Compliance documentation is shared directly with design partners on request.