The demos show an example of my capabilities. Each application and automation solution will be designed from the ground up bespoke to your systems, workflow and environment.
Bespoke AI application · risk simulation
Simultron
10,000simulations / run
The build. A Monte Carlo risk-simulation application — model ten thousand iterations of scenarios on a project and see the full distribution of outcomes, not a single point estimate. Designed, built and shipped end to end as a working web app.
Demo output — the cost-distribution S-curve and risk-driver tornado from a 10,000-iteration run.
What I built
Full-stack simulation app, end to end
10,000 iterations per run in under a minute
AI assisted risk identification of the project risk register and quantification
Able to ingest business cases and project background facts to inform the risk register developed by AI
Why it's here
Proof that AI enables SaaS applications can provide a similar output as a team of advisors, much faster, much lower cost than advisor fees (and current SaaS fees), and of high quality.
Quantitative decision-support software, democratising and making accessible sophisticated risk-quantification techniques for the entire construction sector.
Why it matters
AI is embedded throughout the process in identifying and quantifying the risks, describing them
Saves the user time, improves the quality of the deliverable, ensures they go into the project eyes wide open
Agentic database enrichment
Vagabod
7,986 gyms59 countries enriched
The situation. Vagabod is a real live site that helps travellers find casual gym access anywhere in the world — It solves the problem of gyms not advertising casual membership without a traveller enquiring first. My team of agents reaches out to gyms around the world in multiple languages and enriches a database with information not available from a simple google search or a gyms website.
The live Vagabod landing page — an enriched directory of gyms with casual access across dozens of countries.
What I built
Automated scraping + enrichment pipeline
Casual-membership pricing extraction per gym
Outreach pipeline to inquire about unadvertised casual offerings
Continuous re-enrichment as data drifts
Outcome
7,986 gyms across 59 countries enriched
Directory kept fresh without manual entry
Self-service claim work flow for gym owners to keep their pages current
Why it matters
The same agentic enrichment applies to any large, complex dataset
over 10 Automated workflows & always on, running 24/7 on local LLMs hosted on private infrastructure, for free
The situation. A full virtual top tier consulting engagement team with embedded IP and skills designed to automate the production of advice and deliverables, iteratively reviews agents' work in line with a hierarchy, and produces high-quality, client-ready deliverables.
What I built
A swarm of general purpose agents that can advise on a range of engagement deliverables including transaction management, business cases, transaction evaluation and execution.
Each agent or 'virtual team member' is specialised with their own set of skills such as the commercial lead, financial lead and reports (including financial modeller agent) to avoid bloating context windows and enhance performance
Iterative, proactive goal based loop engineering and review to produce polished products
One single point of contact - the engagement director acts as an orchestrator and tasks the swarm, reviews every deliverable before it is called 'done' and goes out the door.
Outcome
Robust, insightful advice
Faster, more consistent processing
Economical use of AI with budgets and right model for the right job
Why it matters
Leveraging LLMs in a harness to end to end produce deliverables that would have previously cost millions of dollars
Superior quality to simply using a chatbot frontier model or home made skills, iterative review and embedded IP, orchestration and review loops combat hallucination and quality issues
Client ready deliverables saving you time and money. The full suite of deliverables over engagement.
Agent swarm · end-to-end property transactions
Keystone
95 minfull BTR decision memo
The build. A private institution, run by AI. Keystone is a bench of specialist property agents — originate, underwrite, design, operate — orchestrated by a single steward that turns a plain-language question into a governed, decision-ready answer. Seventeen years of property advisory knowledge, encoded as skills and run end to end across the asset lifecycle. Illustrated on a representative A$5B property empire.
From the Keystone deck — an illustrative institution built on representative, non-client data.
What I built
A steward that turns plain-language asks into orchestrated work across the agent bench
