Intelligence layer · native to Cortrova

Subsidiary of Viceroy NM

Trunnion AI

The native AI engine embedded into every module, dashboard, and workflow within Cortrova. Not a bolt-on. The nervous system of the entire platform.

Render 020 · Live System Stats

/trunnion-ai/system-status

Agents 0 AI agents native

Tiers 3 Hierarchical layers

Engines 6 Analysis engines

Analyzers 0 Domain analyzers

Coverage 0 Departments

All systems operational Latency < 1.2s Synthesis cadence · 30s //

Architecture

Three-tier agent hierarchy.

A bottom-up intelligence pipeline. Each tier feeds the next, creating a comprehensive operational intelligence network across all 17 departments.

Render 021 · Agent Hierarchy

/trunnion-ai/agent-hierarchy

// Tier 01

Department Agents

3 agents × 17 departments

Three specialized agents per department covering anomaly detection, recommendation generation, and KPI reporting. The foundation tier of the intelligence pipeline.

Analyst Anomaly detection on dashboard load ~8K ctx

Advisor Per-record recommendations ~4K ctx

Reporter KPI summaries every 30 minutes ~6K ctx

// Tier 02

Division Aggregators

Cross-department synthesis · 30 min cadence

Seven divisions synthesize insights across related departments, surfacing cross-functional patterns invisible to single-department analysis.

OperationsProduction and shop floor~10K

SafetyRegulatory and environmental~10K

FinanceProcurement and accounting~10K

Supply Ch.Logistics and inventory~10K

PeopleHR and facilities~10K

CommercialSales and customer service~10K

TechnologyIT, OT, and R and D~10K

// Tier 03

Executive Agents

5 specialists + 1 synthesizer

Five executive specialists each carry 12K context. The Synthesizer combines all into a unified C-suite briefing at 16K context with 4K output.

StrategyLong-horizon planning~12K

OperationsCross-division performance~12K

PeopleWorkforce and culture~12K

RiskCompliance and exposure~12K

MarketIntelligence and signals~12K

SynthUnified C-suite briefing~16K

Tier 01 → Tier 02 → Tier 03 → C-suite briefing

// 65 agents total Bottom-up synthesis pipeline 17 departments · 7 divisions · 6 executives //

Analysis Engines

Six engines.
One intelligence fabric.

Running every 6 hours and on demand, querying 90 days of data across all 17 departments. The mathematical core under the agent hierarchy.

Render 022 · Analysis Engines

/trunnion-ai/analysis-engines

Engine / 01

Predictive Alert

Linear regression with R-squared scoring. 7 and 30 period SMA deviation with Z-score anomaly detection at ±1.5 sigma. Threshold proximity alerting within 10%. Severity and confidence scoring on every alert.

Cadence · 6hSigma ± 1.5

Engine / 02

Root Cause Analysis

Cross-department correlation detection. When quality drops, traces to overdue maintenance, supplier changes, staffing shifts. Connections invisible in siloed systems become evidence.

Window · 90 dayCross-domain

Engine / 03

Prescriptive Actions

60+ actionable templates with real metric values, effort and cost estimates, historical effectiveness tracking, and projected downstream KPI impact. Owners and timelines included.

Templates · 60+With ROI

Engine / 04

Scenario Modeling

50+ pre-seeded causal relationships with impact coefficients, confidence scores, and lag periods. Model interventions before committing resources.

Causal · 50+What-if

Engine / 05

Resource Optimization

Multi-issue prioritization scoring urgency, impact, availability, and cross-department dependencies. Output is a prioritized action plan with assigned owners.

Multi-factorOwner-tagged

Engine / 06

SPC Anomaly Detection

Continuous Cpk and Cp monitoring. 1 sigma drift detection. 3 sigma out-of-control flagging. Catches drift before defective product ships. Runs 24x7 with zero manual chart review.

24x7Cpk continuous

// 6 engines · 1 fabric 90-day rolling window Runs every 6h + on demand //

Orchestration

Automated intelligence on a schedule.

Four cadences. Always running. The clockwork that keeps every dashboard, alert, and briefing fresh without anyone hitting refresh.

Render 023 · Orchestration Schedule

/trunnion-ai/orchestration

Cadence Action Tier

Every 15 min Department health scores computed for all 17 departments · 0 to 100 composite scoring Tier 01

Every 30 min Reporter agents and Division aggregators generate cross-department insights and KPI summaries Tier 01 + 02

Every 6 hours Full analysis pipeline runs: all 10 domain analyzers, cross-domain correlation, 6 engines, predictive alerts Engines

On demand Executive briefings, Narrator conversations, mock audits, ad-hoc analysis on user request Tier 03

// 4 cadences Always-on synthesis Zero manual refresh required //

Security

Multi-layer AI governance.

Every agent request flows through a seven-stage governance pipeline before reaching an LLM. No exceptions.

Render 024 · Security Pipeline

/security/ai-governance

01 Team Guard Caller authenticated

02 Kill Switch Global, division, team

03 Rate Limit Per-user, per-team RPM

04 Budget Check Daily token budget

05 Scope Validate Data boundary check

06 LLM Call Claude or GPT

07 Audit Log Non-blocking

Control / 01 Kill Switches

Three levels at global, division, and team. Instant shutdown with zero caching delay.

Control / 02 Scope Registry

Declarative data boundaries. Safety agents cannot query Finance models.

Control / 03 Rate Limiting

Per-user, per-team RPM via Redis. Default 60 RPM, configurable up or down.

Control / 04 Token Budgets

Daily budget per team. Warning at 80%. Hard stop at 100%.

Control / 05 API Key Mgmt

Per-team encrypted keys with rotation policy and per-team model allowlists.

Control / 06 Audit Trail

Every agent action logged. Non-blocking. Fully queryable from the audit dashboard.

// 7-stage pipeline · 6 controls Zero exceptions Audit log · non-blocking //

Full Security Details →

Get started

See Trunnion AI in action.

Watch 65 agents analyze real operational data in a personalized walkthrough tuned to your industry.

Request a Demo → View All Features

FAQ

Trunnion AI questions.

Is Trunnion AI included or an add-on?

Fully included. All 65 agents, 6 analysis engines, and signature capabilities are included at no additional cost. There is no separate AI tier.

Can I disable AI for specific departments?

Yes. Three kill switch levels at global, division, and team scope, with instant effect and zero caching delay. Granular controls let any department disable AI assistance independently.

What LLM models does it use?

Anthropic Claude and OpenAI GPT via a provider abstraction layer. Different departments can use different models simultaneously based on task requirements and cost optimization. Model allowlists are managed per-team.