Arrochar Labs
ASG™, the Arrochar Superintelligence Guardrails
The pathway to scaling AI agents across the enterprise, safely.
Gold-standard safety and audit guardrails that take you from a handful of pilots to agents at scale, tailored to the way your people work, with better user satisfaction and designed to lower capex and opex.
Arrochar Labs builds AI products that make the AI inside any organisation visible, governed, and accountable. Every one is a full AI-agentic experience: agentic, event-driven, with human-in-the-loop controls. The governance rails are standard, with assurance that is reused rather than rebuilt. The services that run on them are built to measure, and each is assessed on its own.
ASG™, the Arrochar Superintelligence Guardrails
One suite. One accountable lifecycle.
ASG™ is the whole product family run as one complete suite of AI governance agents: the safety and audit guardrails that let you scale AI agents across the enterprise, safely, and retire the software platform giants one service at a time, with every agent visible, governed, and accountable. Two products are the rails every deployment runs on. The rest are built to measure on top of them, and the Industry Agent Starter Packs are the same suite packaged for one industry.
All products run on secure, scalable AWS cloud infrastructure, with data residency in the region you operate in.
AIG Sentinel
Demo availableAI governance for any organisation
The AI governance agent for organisations anywhere. AIG Sentinel finds the AI you already use, assesses each use against the frameworks that apply to you, and produces an audit-ready evidence trail - so any organisation can manage its AI governance with confidence.
Request demo accessOrbit Roadmaps
Demo availableAssess AI opportunities, build the roadmap
Orbit assesses where AI can add value across your organisation, then builds the AI roadmap to capture it - turning a short brief into a phased, costed, dependency-mapped roadmap with risk profiles and stakeholder engagement, and tracking delivery against it.
Request demo accessMeshbone
Available nowThe control plane for AI agents
Meshbone lets an organisation run and govern all its AI agents and AI services from one control plane, the way a service mesh manages microservices: what you run, who owns it, what it costs, and whether it still meets policy - from first request to retirement.
Request demo accessagentVIKING™
22 of 27 controls run todayDetect, prevent and correct rogue AI agents
agentVIKING™ is the set of rogue-agent controls every Arrochar Labs product enforces on every agent it runs: directive, preventive, detective and corrective. An agent that acts beyond its delegation, outside its charter, on data it was not cleared for or on an instruction it found in a document is stopped, seen and put right. Written as policy in AIG Sentinel, enforced at run time by the gateway in front of any agent's model calls, ours or not, carried inside every Nanolitte template and evidenced on one tamper-evident chain. 22 of the 27 controls run today; the rest are in development.
Nanolitte
In developmentThe build engine for every governed service
Nanolitte is the build engine of the suite. It keeps every service as a BPMN 2.0 process, compiles each task into a governed agent and each person-step into a hold for approval, and carries the agentVIKING™ guardrails inside every template it produces. The Industry Agent starter packs, MeshGov and MeshFin among them, are built from Nanolitte templates; it has no catalogue of services of its own. Build with Nanolitte, deploy and govern on Meshbone.
Request demo accessMeshGov
Demonstration runningTransform government with AI automation
Governed AI agents for the operations of government: grants, cases, regulation, procurement and the rest of the twenty-one services, automated to save public money, with sovereign data owned by government. Modern interfaces for the officers who run government and the citizens who use it, with shared rails that cut corporate risk and cost together. Any portfolio, in any country, built on Meshbone with the highest security architecture we hold any product to.
Request demo accessMeshFin
In developmentRun the bank on governed AI
Governed AI agents for the operations of a bank, a mutual or a non-bank lender: onboarding, lending, payments, complaints, hardship, financial crime, compliance and the rest of the twenty-four services, mapped service by service to the BIAN Service Landscape. Agents carry the work; a named person takes every decision with legal, financial or regulatory effect. The financial-services sibling of MeshGov, on the same rails. In development: every service is scoped and none runs for a customer yet.
MeshAero
Work in progressRun the airline on governed AI
The suite packaged for the operations of an airline, an airport or a maintenance organisation: schedules and slots, crew records, airworthiness and maintenance, the safety management system, ground handling, passenger disruption and refunds, security identification and the reporting a civil aviation regulator expects, as governed AI agents that never dispatch a flight or release an aircraft to service.
MeshVolt
Work in progressRun the utility on governed AI
The suite packaged for the operations of an energy utility: connections and metering, outage and fault management, asset maintenance, works and contractors, market settlement, customer billing and hardship, safety incidents and the reporting an energy regulator expects, as governed AI agents that never take a real-time control action on the grid.
MeshAgentDesigner
In developmentShape the starter pack to the way your organisation works
Every Industry Agent starts as a standard pack: a catalogue of services, each with its agents, its decision gates and its evidence records, identical for every customer. MeshAgentDesigner turns that pack into yours. It can watch how your people do the work today and propose the changes, read the process you drew in Lucidchart or another drawing tool, or take the changes you configure by hand. Every change is versioned, held for a named owner where it carries weight, and reported in the three columns every build is measured in.
Databallast
In developmentData ready for AI, at scale
Databallast gets your organisation's data ready for AI at scale: a governed business glossary, metadata and lineage, a logical business information model, data maturity assessment, and defined data products - with every dataset certified AI-ready under DAMA-aligned governance.
Request demo accessExecdive
Demo availableAI-driven analytics for executives
An AI analytics agent that lets executives ask questions of their organisation's data in plain language. Execdive works across your data lake, brings in external data where it adds context, and answers with a written synopsis plus charts in whatever form you ask for.
How it fits together
A complete suite of AI governance agents. An architecture that makes them accountable.
Each part of ASG™ does one job, in order, and hands the next part something it can check. That is the difference between building AI agents and running them so an auditor, a regulator or a minister can see what happened.
Before the first build, Orbit Roadmaps finds where AI creates value and profiles the risk.
After it, Execdive turns what the chain recorded into plain-language insight for executives.
Policy, enforcement, templates, data, controls, evidence. Then services.
This describes the architecture, not what every deployment runs today: 5 of the 27 agentVIKING™ controls are still in development.
- AIG Sentineldefines policy
- Meshboneenforces policy at run time
- Nanolittecarries the controls into reusable service templates
- Databallastcontrols what data the agent may access
- agentVIKING™is the control set across every agentAcross every step
- Evidence chainrecords what happened
- MeshGovuses the whole thing for government services
- MeshFinuses the whole thing for banking and financial services
How every deployment is structured
Standard rails. Made-to-measure services.
The governance rails are ready-made and the same for every customer, so their assurance is reused: assessed once, carried into every service, while each service is still assessed on its own. The services that run on them are built to fit the way your organisation actually works. You get software that fits, and it is still governed.
Governance rails
Ready-made. Identical for every customer.
AIG Sentinel sets the policy and holds the evidence. Meshbone enforces it at run time on every agent and service. agentVIKING™ is the set of rogue-agent controls the rails enforce on every agent they run. Because the rails are the same everywhere, an assessor who has assessed them once need not assess them again for the next service. Each service still carries its own assessment of its purpose, data, risks, integrations, rules and model use. The rails make that assessment smaller. They do not replace it. The rails subscription is the derisking spend: the controls, the monitoring and the evidence that the AI tools themselves do not include, paid for once and reused by every service rather than rebuilt around each one.
Service templates
Productised. Drawn from the catalogue.
Nanolitte is the build engine: every template is a BPMN 2.0 process it compiles into governed agents, with the guardrails inside. Grants, cases, permits, procurement, records, assets: each starts from a template that already runs on the rails, with intake, assessment, decision and record built in. A new service begins most of the way there rather than from a blank page.
Made to measure
Built for you. Kept small by design, measured on every build.
Your rules, your data model, your thresholds and your approval chain, configured and coded onto the template. This is where the hard part of any regulated service lives: the legislation, the exceptions, the delegations and the systems you keep. So we scope that share in three columns before a build starts, standard, configured and coded, and report it in the same three when it ends. Anything a second customer could use goes back into the template.
Run and support
Ongoing, on the rails.
Hosting, monitoring, policy re-attestation and upgrades, in your region. After go-live your own team changes a threshold or a form within guardrails, without waiting on us.
The rule
No made-to-measure build without the rails underneath it. That is the rule that keeps a custom service governed, upgradeable and provable, rather than a fork somebody has to maintain by hand forever.
Organisations are already building instead of buying
The build wave has arrived. Most of it is ungoverned, half of it never reaches production, and the barrier people name is the one the rails remove.
Source: Retool, The Build vs. Buy Shift (2026), a survey of 817 customers and builders
What you are left with
Guarded by agentVIKING™
Every agent, ours or yours, under the same four classes of control
Directive, preventive, detective and corrective. Written in AIG Sentinel, enforced at the gateway in front of the model, carried in every Nanolitte template, evidenced on one chain.
See the controlsDirective
What the agent is allowed to do, as data.
Preventive
What stops it before it acts.
Detective
How we know when it goes wrong.
Corrective
How it is stopped, reversed and fixed.
One architecture, every product
A full AI-agentic experience
Every Arrochar Labs product is built the same way: agentic, event-driven, and governed with human-in-the-loop controls, so AI does the work while your people stay in control.
Agentic
AI agents do the work, end to end. Every product runs on autonomous agents that take action across your processes, not just a chat box, but software that actually gets the job done.
Event-driven
It reacts in real time. Products respond to what's actually happening (new data, a policy breach, an incoming request, a change in your estate) instead of waiting on a batch run or a person to press go.
Human-in-the-loop
Your people stay in control. Approval gates, oversight and a tamper-evident audit trail keep every agent accountable, so automation never runs unchecked.
The Accountable AI Framework
From opportunity to accountable value. Our products, together ASG™, the Arrochar Superintelligence Guardrails, form one lifecycle for adopting AI you can trust - discover where it helps, govern it, build agentic services, operate them, ground it all in trusted data, and turn it into decisions. Two of them, AIG Sentinel and Meshbone, are the rails every deployment runs on. The rest are built to measure on top of them.
Find where AI creates real value.
ExploreMake every AI safe, compliant and provable.
ExploreThe build engine: draw the service, get governed agents.
ExploreRun and connect AI across the organisation.
ExploreGround AI in data it can trust.
ExploreTurn AI into executive insight and action.
ExploreOur software moves you from → to
Most organisations have AI they can't fully see, govern, or trust. Our products close that gap - turning scattered AI experiments into governed, accountable, production-grade systems.
How the products work together
Two products are the rails: AIG Sentinel sets the policy and Meshbone enforces it, with agentVIKING™ guarding every agent they run. The rest are built to measure on top of them: Orbit finds the value, Nanolitte builds the services, Databallast grounds them and Execdive turns them into decisions. One connected suite, from the AI you can't see today to a governed, production-grade stack.
Discover
See every AI you run.
- *AIG Sentinel surfaces the AI already in use across your organisation - including the shadow AI hiding in everyday tools.
- *Orbit reads your strategic outcomes and current technology, data and processes to find where AI adds the most value.
- *You start from a complete, accurate picture - not a blank page.
Govern & build
Assess, approve, and make it AI-ready.
- *Each use case is assessed against the frameworks that apply to you, routed through approval gates, with a tamper-evident evidence trail.
- *Orbit turns the gaps into a phased, costed roadmap of work packages with current and target-state architecture.
- *Services are built to measure on the rails: a Nanolitte template, configured to your rules and data model, with the custom share scoped before the build and reported after it.
- *Databallast builds the governed, AI-ready data asset - glossary, catalog, information model - and Meshbone stands up the control plane your AI services depend on.
Operate & improve
Run it with confidence.
- *Meshbone runs and governs every AI agent and service from one control plane - what you run, who owns it, what it costs, and whether it still meets policy.
- *agentVIKING™ guards every agent as it runs: outside its charter it is blocked, drifting it is flagged, and the kill switch stops it.
- *Continuous monitoring, drift detection and annual re-attestation keep approvals current as your environment changes.
- *Execdive turns the data into plain-language answers for the executives who need them.
Outcomes you should expect
Every Arrochar Labs product is built to deliver measurable, lasting results - not just dashboards.
AI infrastructure is now commoditised. Trustworthy AI isn't.
Our products make the AI you already run visible, governed, and accountable.
Built secure from day one.
Every Arrochar Labs product is built so the AI inside your organisation is something you can trust - built to align with ISO/IEC 42001 (AI management), ISO 27001, SOC 2, the OWASP LLM Top 10, and the privacy and assurance regimes that apply to you, wherever you operate.
Data sovereignty
Data-residency options in the region you operate in. Your data is never used to train external models.
ISO/IEC 42001 by design
Built to align with the international standard for AI management systems - governance designed into the product, not bolted on.
Security hardened
Built to recognised baselines - the CIS Controls, the NIST Cybersecurity Framework, and the ASD Essential Eight.
Privacy by design
GDPR, CCPA, and the privacy regimes that apply to you - embedded in the product architecture.
AI-specific safety controls
Prompt injection, hallucination, and data-poisoning defences, aligned to the OWASP LLM Top 10.
Audit-ready by default
Structured logging, drift detection, tamper-evident evidence, and audit-ready dashboards in every product.
We don't sell hype. We build the products that make AI actually work.
Most AI initiatives fail not because of the technology, but because the AI is invisible, ungoverned, and stuck in pilots. ASG™ is the pathway to scaling AI agents across the enterprise, safely: gold-standard safety and audit guardrails, and the tailored, lower-cost way off the software platform giants.
Scale safely
Every agent visible, governed and accountable, from pilot to production. AIG Sentinel writes the policy, Meshbone enforces it at run time, and the evidence chain records what happened.
See the rails every agent runs onGold-standard guardrails
agentVIKING™ on every agent, with a full audit evidence chain. 22 of the 27 controls run today, built to align with ISO/IEC 42001, ISO 27001, SOC 2 and the OWASP LLM Top 10.
Read the controls list and the standards mapThe way off the platform giants
Tailored to your organisation rather than a vendor's roadmap. Users get a better experience, you spend less upfront, and the design is for operating costs that fall over time.
How a starter pack is shaped to youDeep expertise across regulated industries
We understand the procurement rules, compliance frameworks, and operational constraints of the sectors we work in. Each one has an Industry Agent Starter Pack.
Public-sector organisations we have worked with
- Transport NSW
- Snowy Monaro Regional Council
- Port Macquarie Hastings Council
- NSW Reconstruction Authority
- Kempsey Shire Council
- Byron Shire Council
- NSW Dept of Finance, Services & Innovation
- Newcastle Airport
- DPIE Water
- NSW National Parks and Wildlife Service
Delivered as Arrochar Consulting engagements in ICT strategy, data, architecture and smart-place work. The software products are newer than this record, and we say so on the hard questions. Read the case studies →
IoT Global Awards
Recognised globally for AVARI platform - multi-source IoT data fusion for real-time analytics.
Multi-Jurisdictional Panels
Pre-qualified across SCM0020, QLD 1303B, and LGP panels.
Claude Partner Network
We build on Claude, with governance and assurance built in, as a member of Anthropic's Claude Partner Network.
Common questions
Partnerships
Built on the platforms our clients run on.
We hold formal AI partnerships and program memberships with the cloud, silicon, and model providers powering enterprise AI today, so our products are supported and roadmap-aligned.
AWS AI Partner
AWS Partner Network - AI Specialisation
Active partner
Google Cloud AI Partner
Google Cloud Partner - AI/ML
Active partner- Claude Partner Network
Claude Partner Network
Member of the Claude Partner Network
Program member 
NVIDIA Inception Program
Member of the NVIDIA Inception program
Program member
© 2025 NVIDIA, the NVIDIA logo, and NVIDIA Inception are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. All other trademarks are the property of their respective owners.
See what our products can do for you.
Tell us which product you're interested in and we'll walk you through it against your own use cases.
A demo, not a pitch
We show the product end to end against your own use cases - no account managers, no pressure. You decide in your own time.
Fast to value
AIG Sentinel is live in 24 hours. See governed, audit-ready AI without a long procurement cycle.