AI Consultant & Automation Engineer.
~$1M/month in automated financial operations.
The same framework, built for your business.
Production-tested AI architecture for CEOs and SMEs. You pay when milestones are met. Nothing more.
Built in Production. Not Theory.
I spent three years building the AI operations infrastructure for a fast-growing property management company. Payment bots, acquisition pipelines, onboarding automations. Five operational teams depending on systems I designed and shipped.
That pressure is where I developed the 5-layer agent framework. Behavioral rules in markdown, factual knowledge in semantic search, plan-approve-execute safety gates on every consequential action. It's now running across multiple production systems.
I productized that framework into a consulting engagement. The same architecture that processes ~$1M/month in automated financial operations is the one I build for CEOs and SMEs who are ready to run on AI.
I specialize in the Anthropic stack. If your ideal future runs on Claude, let's talk.
Automated Financial Volume
~$1M
per month, all streams
Companies Automated
5
clients served
API Integrations
Any REST API
if it has docs, we connect it
The 5-Layer Agent Framework
I built this under real operational pressure. Five layers that turn an AI tool into a system your business can actually rely on. The architecture that handles $550K/month in financial operations is the one I build for you.
Persona & Role Governance
The agent's identity, authority boundaries, and hard limits. It knows what it can and can't do before any task starts. No scope creep, no unauthorized actions.
Skills & Tool Registry
Discrete, callable capabilities. Each skill is named, scoped, and independently deployable. You add new features without touching the core agent.
Hybrid Memory
Behavioral rules in markdown (always loaded). Factual knowledge in semantic search (hippo-memory, SQLite-backed). The agent gets smarter the longer it runs.
Plan-Approve-Execute Gates
Every consequential action gets a plan you review first. The agent proposes. You decide. No writes, sends, or mutations happen without your explicit approval.
Orchestration & Integration
API connectors and event triggers that plug the agent into the tools your business already uses. HubSpot, QuickBooks, Guesty, Xero, and more. It acts where work actually happens.
I specialize in the Anthropic stack. If your ideal future runs on Claude, let's talk.
Book a strategy call →Services
Every engagement is milestone-gated. You see the work before the next phase starts.
AI Architecture Audit
Not sure where to start? Fill out the async discovery form and I'll map your current setup, identify the gaps, and give you a prioritised build roadmap. No call required.
AI OS Build
I design and build your full AI Operating System. Agents with governance, memory, and orchestration, all connected to the tools your business already runs on. Structured on GitHub so you can see and run every part of it.
Workflow Automation
Got a manual process eating hours every week? I build cloud-deployed Python scripts, Base44 apps, or n8n/Make flows to take it off your plate. Scoped tight, delivered fast.
Ongoing Support
Keep building after delivery. Pick what you need from the menu below. No monthly commitment.
- › Monthly system report + async priority review ($150/mo)
- › Skill iteration, one existing skill, 2 feedback rounds ($200)
- › New skill development, greenfield ($350+)
- › New MCP or API integration (quoted per integration)
- › Scheduled task or automation job (quoted per job)
Not sure which engagement fits? Let's figure it out in 30 minutes.
Book a Free Strategy CallSuccessful Engagements
Real engagements. Verified figures. These systems run in production today.
What Clients Say
Testimonials in progress. References available on request.
Book a Strategy Call
A free 30-minute call to discuss your AI goals. No pitch. I'll tell you honestly whether I'm the right fit and what I'd build first.