How we work
We don't hand you a platform and disappear.
A framework alone doesn't get an agent into production — the deployment work does. Every engagement pairs our five-principle framework with people who've done this integration before — yours doesn't have to be the first attempt.
The methodology
Diagnose. Architect. Deploy. Govern.
The same four phases, whether it's a two-week pilot or a multi-team rollout.
Diagnose
We sit with the team actually doing the work — not just the leadership sponsor — and map where an agent would create real leverage versus where it would just move the bottleneck. You get a written assessment either way, including a straight answer if we don't think this is a fit yet.
Architect
Before anything touches production, we design the agent's identity, its permission scope, and the specific actions that require a human sign-off. This is the document your security and compliance teams review — and the one that gets projects approved on the first pass.
Deploy
We integrate directly against your live systems — your CRM, your case management tool, your ERP — with the human checkpoint active from day one. The first workflow goes live in weeks, not after a quarter-long platform migration.
Govern
Your team takes ownership of the policies. We stay involved to monitor performance, tune the governance rules, and expand scope only once the audit trail has earned it — not on a fixed rollout schedule.
Where to start
Different roles walk in with different questions
Engagement tracks
Four ways to work with us
Diagnostic Engagement
A fixed-fee, fixed-timeline workflow audit. You get a written roadmap and a risk assessment — with no obligation to continue.
- 1–2 week engagement, fixed fee
- Workflow mapping + automation roadmap
- Best for: teams evaluating whether this is worth doing at all
Deployment Engagement
Project-based build and integration. We architect and deploy the first one to three agent workflows into your live systems.
- 4–8 week engagement, scoped fixed fee
- Full architecture, integration, and go-live
- Best for: teams ready to put an agent into production now
Managed Operations
Ongoing subscription. We monitor, tune, and expand your deployed agents so your team doesn't have to run it as a side project.
- Monthly subscription, cancel anytime
- Ongoing tuning, monitoring, and policy updates
- Best for: teams without a dedicated AI operations function
Security & Governance Audit
From AU$7,500 fixed fee
A standalone review of an existing AI deployment — including ones we didn't build — against the same five-layer standard.
- 2–3 week engagement, fixed fee
- Written findings against identity, governance, and audit gaps
- Best for: teams that already deployed AI tools and need to know their exposure
The audit, in detail
What AU$7,500 actually buys you
A written, evidence-based review of AI tooling already running in your business — whether we built it or not — checked against the same five-layer standard (Identity, Orchestration, Knowledge, Governance, Audit) we use on every deployment.
Discovery
We inventory every agent, script, or AI-powered process currently touching your systems — including the ones nobody remembers approving. Access, credentials, and data flows get mapped as they actually exist, not as the org chart says they should.
Review
Every identity, permission scope, and approval gap gets checked against the five-layer standard. Findings are graded by actual exposure — a shared credential with write access doesn't get filed the same as a read-only integration.
Findings & readout
You get a written report — findings ranked by exposure, and what fixing each one would take — walked through live with your team, not left as a PDF in a drawer. Remediation is optional and scoped separately; the audit doesn't require hiring us again.
Included
- Full inventory of AI tools and agents with access to your systems
- Identity and permission-scope review against least-privilege practice
- Written findings report, ranked by exposure, not just a checklist
- Live readout session with your team, not just an emailed report
Not included
- Remediation or implementation work — scoped and quoted separately if you want us to fix what we find
- Penetration testing or application code review — different discipline, happy to refer out
- A formal compliance certificate — we're early-stage and don't issue or claim one