Decision owner
NOC engineer
The named role remains accountable for the consequential decision.
Telecommunications
Designed for NOC engineer
Cut mean time to repair in the network operations centre by correlating alarms with topology and change records, retrieving the matching runbook, and requiring engineer approval before any automated remediation touches the network.
Contract TEL-01 · primary aggregate NetworkIncident · mode loading
Decision brief
This brief starts with the decision, evidence, and accountable owner. The guided workflow then shows how the recommendation was produced.
The evidence and control points that shape this recommendation.
Decision owner
NOC engineer
The named role remains accountable for the consequential decision.
Governed evidence
6 record types
Structured records and cited text ground the recommendation.
Workflow
5 controlled stages
Each transition records its input, output, status, and audit context.
Human controls
1 approval gate
Approval gates prevent the copilot from making consequential decisions.
At 02:10 a metro region raises 380 alarms; the copilot correlates them to a single aggregation router, links the onset to a configuration push 14 minutes earlier, retrieves the rollback runbook, predicts 26,000 subscribers restored, and the NOC engineer approves the rollback which clears the incident in nine minutes..
Cut mean time to repair in the network operations centre by correlating alarms with topology and change records, retrieving the matching runbook, and requiring engineer approval before any automated remediation touches the network.
Executed change with rollback plan and incident update gives NOC engineer the evidence needed to begin the final governed business outcome, with the decision trail retained for review.
Step 1 of 5
| Field | Value |
|---|
| Field | Value |
|---|
The output appears here immediately. More technical evidence remains available below.
The demonstration keeps deterministic controls, AI assistance, and approval separate.
Source systems enforce identity, workflow order, policy filters, and record updates.
Retrieval and generation assemble relevant evidence into a cited recommendation or draft.
NOC engineer verifies material claims and approves the consequential action.
Run the stages in order before using the Copilot. Every stage explains what happens, why it matters, the evidence it uses, and the complete API request and response.
Automated stage
What happens: Fault management integration performs ingest and deduplicate alarms using Alarm stream from network elements and produces Deduplicated alarm set per region. The transition is deterministic and does not require autonomous model approval.
Why it matters: Deduplicated alarm set per region gives Correlation agent the evidence needed to begin Correlate alarms into a service incident, with the decision trail retained for review.
Loading relevant sample records.
Ready to run.
POST /api/domain/workflow/ingest-and-deduplicate-alarms
Run this API to see its formatted outcome.
| Field | Value |
|---|
| Field | Value |
|---|
Automated stage
What happens: Correlation agent performs correlate alarms into a service incident using Alarms plus topology dependencies and produces NetworkIncident with service impact. The transition is deterministic and does not require autonomous model approval.
Why it matters: NetworkIncident with service impact gives Causation agent the evidence needed to begin Test change and maintenance causation, with the decision trail retained for review.
Loading relevant sample records.
Ready to run.
POST /api/domain/workflow/correlate-alarms-into-a-service-incident
Run this API to see its formatted outcome.
| Field | Value |
|---|
| Field | Value |
|---|
Automated stage
What happens: Causation agent performs test change and maintenance causation using Recent change records in the blast radius and produces Suspected cause with supporting change evidence. The transition is deterministic and does not require autonomous model approval.
Why it matters: Suspected cause with supporting change evidence gives Runbook retrieval agent the evidence needed to begin Retrieve and adapt the runbook, with the decision trail retained for review.
Loading relevant sample records.
Ready to run.
POST /api/domain/workflow/test-change-and-maintenance-causation
Run this API to see its formatted outcome.
| Field | Value |
|---|
| Field | Value |
|---|
Automated stage
What happens: Runbook retrieval agent performs retrieve and adapt the runbook using Fault signature and element type and produces Ordered diagnostic and remediation steps. The transition is deterministic and does not require autonomous model approval.
Why it matters: Ordered diagnostic and remediation steps gives NOC engineer the evidence needed to begin Approve and execute remediation, with the decision trail retained for review.
Loading relevant sample records.
Ready to run.
POST /api/domain/workflow/retrieve-and-adapt-the-runbook
Run this API to see its formatted outcome.
| Field | Value |
|---|
| Field | Value |
|---|
Human approval required
What happens: NOC engineer performs approve and execute remediation using RemediationProposal with predicted impact and produces Executed change with rollback plan and incident update. A named human approver must review the evidence before the transition completes.
Why it matters: Executed change with rollback plan and incident update gives NOC engineer the evidence needed to begin the final governed business outcome, with the decision trail retained for review.
Loading relevant sample records.
Ready to run.
POST /api/domain/workflow/approve-and-execute-remediation
Run this API to see its formatted outcome.
| Field | Value |
|---|
| Field | Value |
|---|
Run a stage to see its explanation, evidence, full request, and full response.
Ask a question and receive an answer grounded in retrieved NetworkIncident evidence.
Select a starter question or enter your own.
Loaded live from /api/config, /api/domain/model, and
/api/domain/sample.
networkIncidents/tenantId + /networkRegionIdAt 02:10 a metro region raises 380 alarms; the copilot correlates them to a single aggregation router, links the onset to a configuration push 14 minutes earlier, retrieves the rollback runbook, predicts 26,000 subscribers restored, and the NOC engineer approves the rollback which clears the incident in nine minutes.
| Container | Role | Hierarchical partition key | Document types |
|---|---|---|---|
networkIncidents |
Primary | /tenantId + /networkRegionId |
NetworkIncident, NetworkAlarm, TopologyRelation, ChangeRecord, RemediationProposal |
networkRunbookLibrary |
Reference | /tenantId + /faultSignatureCode |
RunbookProcedure |
| Type | Container | Partition key | Identifier pattern | Embedding fields |
|---|---|---|---|---|
NetworkIncident |
networkIncidents |
/tenantId + /networkRegionId |
incident::{networkRegionId}::{incidentId} |
impactSummaryVector |
NetworkAlarm |
networkIncidents |
/tenantId + /networkRegionId |
alarm::{networkRegionId}::{alarmId} |
None |
TopologyRelation |
networkIncidents |
/tenantId + /networkRegionId |
topology::{networkRegionId}::{elementId} |
None |
ChangeRecord |
networkIncidents |
/tenantId + /networkRegionId |
change::{networkRegionId}::{changeId} |
None |
RunbookProcedure |
networkRunbookLibrary |
/tenantId + /faultSignatureCode |
runbook::{faultSignatureCode}::{runbookId}::{stepIndex} |
stepVector |
RemediationProposal |
networkIncidents |
/tenantId + /networkRegionId |
remediation::{networkRegionId}::{proposalId} |
None |
Read alarms and topology for the region inside the correlation window in one partition query.
networkIncidents
single-partition
SELECT * FROM c WHERE c.tenantId = @tenantId AND c.networkRegionId = @networkRegionId AND c.type IN ('NetworkAlarm','TopologyRelation') AND c.raisedUtc >= @windowStartUtc ORDER BY c.raisedUtc ASC
Find changes touching the impacted elements shortly before the fault.
networkIncidents
single-partition
SELECT * FROM c WHERE c.tenantId = @tenantId AND c.networkRegionId = @networkRegionId AND c.type = 'ChangeRecord' AND ARRAY_CONTAINS(@impactedElementIds, c.affectedElementIds[0]) AND c.implementedUtc > @changeLookbackUtc ORDER BY c.implementedUtc DESC
Knowledge-container query against networkRunbookLibrary (partition-aligned to /tenantId + /faultSignatureCode): retrieve the runbook steps whose fault signature matches the correlated incident; the library stores one runbook document per fault signature so the lookup is a single partition read.
networkRunbookLibrary
single-partition
SELECT * FROM c WHERE c.tenantId = @tenantId AND c.faultSignatureCode = @correlatedFaultSignature AND c.type = 'RunbookProcedure' AND c.approvalState = 'approved' ORDER BY c.stepIndex ASC
Show unresolved incidents ordered by customer-impacting priority.
networkIncidents
single-partition
SELECT * FROM c WHERE c.tenantId = @tenantId AND c.networkRegionId = @networkRegionId AND c.type = 'NetworkIncident' AND c.incidentState != 'resolved' ORDER BY c.priorityLevel ASC