Hands-on architect with 25+ years in enterprise software. SaaS APIs, microservices, data platforms and AI systems in C#/.NET — plus a suite of privacy-first local-AI apps I design and ship myself.
I find the opportunity, prove it with working software quickly, then get the organisation to adopt it. I write the code, set the architecture and lead the people.
LLM applications, RAG, agents and governed AI gateways — practical systems people use daily, with PII protection and audit built in.
.NET microservices on Dapr, event-driven design, API and SaaS platforms, observability from day one.
Azure Data Factory, Databricks, Purview and Power BI — and the operating model and culture that make a platform stick.
At Foxit, the CEO, CIO and CCO moved me between units to start new things. These are the ones I'm proudest of.
Proposed the cloud offering and coded the full stack: developer portal, self-service API provisioning, SDK samples, docs and a live API console.
Product Owner and Scrum Master for four engineers. One-week sprints with open reviews for anyone in the company.
Introduced service invocation, pub/sub, state, actors and distributed tracing to a 15-person German/Irish team, with a reference implementation.
Designed and built on Azure, and taught data ownership: intake, prioritisation, classification and department-run data teams.
Connectors for Microsoft 365, Salesforce and Jira; agents; a knowledge graph with tiered memory; and a governed AI gateway with PII redaction.
Aligned Finance teams across China, the US, Germany and Ireland: data migration, integration, training and reporting.
Through my own company I design, build and publish local-AI software — apps that run on your device, keep your data with you, and ship to Windows, macOS, Linux, iOS and Android.
A cross-platform .NET 10 / Avalonia platform of 350+ shared libraries powers 60+ AI apps. 14 products are published on the Microsoft Store, Apple App Store and Google Play, in 16 languages.
llama.cpp, Microsoft Foundry Local (NPU/GPU), ONNX Runtime and cloud models behind one streaming API — with model download, self-update, tool calling and routing.
Microsoft Agent Framework and MCP, hybrid retrieval (vector + BM25 + reranking) and a personal knowledge-graph memory.
Consent-gated, content-free telemetry; signed licences across three app stores; end-to-end encrypted device federation; an OpenAI-compatible enterprise AI Server.
The full detail is in my résumé. Here is the short version.
Designing and building a cross-platform suite of privacy-first, local-AI applications.
SaaS API platform, e-signature product, .NET/Dapr microservices, Azure data platform and internal AI platform. Worked with teams in Ireland, Germany, China and the US.
Grew a bootstrapped consultancy to 20+ engineers serving GSK, Prudential, ThyssenKrupp, TÜV Rheinland, Veolia and Giesecke+Devrient. Every project on time and on budget. Built a mobile GIS and workflow system for 10,000+ ThyssenKrupp Elevator technicians.
Led the .NET re-architecture of EF.com and Englishtown.com (2M users, 100+ countries) across Boston, Hong Kong, Shanghai and India. Cut release effort by about 70%.
Led the prototype that won a HK$30M smart-card tender (V.H. Soft); IBM and Oracle pre-sales and delivery; team lead of 10 developers designing a WfMC-standard workflow engine SDK.
Depth where it counts: C#/.NET daily for over 20 years.
BSc (Hons) Computer Science, First Class — University of Hertfordshire, UK
Higher Diploma, Computer Studies, Distinction — City University of Hong Kong
Champion (twice), Jiangxi Innovation & Entrepreneurship Competition · Top 10, Alibaba Cloud Startup Contest · 2nd Prize, Microsoft Web Services Xcellence Awards · 3rd Prize, Sun J2ME Open
Hiring for a Principal / Staff engineering, architecture or AI platform role? Or want to discuss a project? Send me a message and I'll reply within a working day.
Idea to working software in weeks, not quarters.
Start a conversation