Applied AI Consultant · Solihull, UK
James Farrelly
I turn business problems into practical AI - scoped, shipped, and explained to the people who sign it off.
Open to interesting problems
JamesGPT
Don’t just read a CV. Interrogate it.
Meet JamesGPT - trained on my real CV and answering as me. Ask it anything about my work; if it doesn’t know, it’ll point you to the best way to reach me. It’s the same kind of natural-language interface I build for clients.
JamesGPT · trained on my real CV, answering as me.
01 · About
I sit on the line between the people who need the outcome and the people who build it.
I scope the challenge, define the LLM and agent workflow, write the brief and specification, then guide developers through delivery.
Because I’ve shipped production AI agents and automations across the Microsoft and Azure stack myself, the advice is grounded in what works at scale - and I build governance and security in from the start.
02 · Selected work
Things I’ve built and shipped.
-
01
0 premium licences
Natural-language interface over Power BI
Plain-English chat over reporting data. An MCP server reads the schema, an Azure model writes the query, and AI formats the result - non-technical users self-serve instead of waiting on an analyst.
-
02
Won hackathon
AI research & outreach agent
Takes a list of contacts, researches the matching companies, and turns messy web data into a clean, structured dataset - replacing hours of manual research.
-
03
6+ hrs / month
Verification Automator
A single email triggers validation against a source system and returns a client-ready summary - with nothing manual in between.
-
04
Production-grade
Event-driven reporting pipeline
A file upload turns an unstructured workbook into a structured reporting template - built to production standards with error handling, logging and retries.
-
05
8 → 1 steps
One-click QC tool
Replaced a manual, eight-step, three-hour quality check with a one-click tool - letting the team do more with less and removing a recurring third-party licence cost.
-
06
Forecast-led
Predictive analytics for a major client
Forecasting models turn survey data into a forward view of capital spend - shifting planning from reactive to forecast-led.
03 · By the numbers
04 · Capabilities
Where I’m useful.
- AI & agents
- LLM orchestration with model fallback, prompt engineering, agent workflow design, MCP integrations.
- Microsoft & Azure
- Azure AI Foundry, Azure OpenAI, Copilot Studio, Power Platform, Microsoft Graph API.
- Consulting & delivery
- Requirements gathering, briefs & specifications, solution design, roadmap ownership, live client demos.
- Governance
- Responsible, secure-by-design AI, risk & governance awareness, production-grade delivery.
- Engineering
- Python, JavaScript / Node.js, REST APIs, event-driven workflows.
- Tooling
- Git, Claude Code, Cursor, Windsurf, GitHub Copilot, N8N, the Azure stack.
05 · How I work
A repeatable way to get from problem to production.
-
01
Scope the challenge
I start with the business problem, not the technology.
-
02
Define the workflow
I design the LLM and agent workflow that actually solves it.
-
03
Write the brief & spec
I turn it into a clear brief developers can build from.
-
04
Guide delivery
I steer the build to production - governance and security in from step one.
06 · Recognition & education
- Winner, Morta AI Hackathon
- Distinction, Multiverse Data Fellowship
- BSc (Hons) Quantity Surveying - University of Reading
- Digital Consultant, RLB Digital
07 · Contact
Let’s scope your challenge.
Tell me the business problem - I’ll tell you whether AI is the right tool, and what it would take to ship it.
jamesbernardfarrelly@gmail.com- LinkedIn →
- wafflerai.com →
- Solihull, UK