AI integration & consulting
AI features that do real work inside your software.
LLM features, retrieval, agents, and evaluation. I help you work out what AI changes for your product, where it's worth building, and where it isn't. No vendor lock-in by default, and no AI features for the sake of having AI features.
The useful AI work in most businesses is unglamorous: drafting, classifying, checking, and summarizing inside a workflow people already follow. It works when it sits in the tool your team uses and when someone has measured that the output is good enough.
I build those features and the evaluation that keeps them reliable. If the answer for your case is that AI will not help yet, I will tell you that before you spend the budget.
- OpenAI · Anthropic
- Vercel AI SDK
- RAG
- Eval
- Agents
What you get
- A written assessment of where LLMs help and where they're a distraction.
- Prototype features (retrieval, structured output, classification, agents) you can put in front of users.
- An evaluation pipeline, so you know if quality regresses with the next model swap.
- A provider strategy: when to pin to a vendor and when to stay portable.
When it fits
- Your team wants to ship AI features but isn't sure which ones.
- You've shipped an LLM feature and don't trust the output.
- Your board is asking about AI strategy and you want a grounded answer.
In production
Where I’ve done this before.

TELUS · Internal tooling
TC Tools
AI-assisted content-ops toolkit over dotCMS
Pre-publish toolkit and content-ops console for the editors and operations staff who keep TechCentral honest.
TELUS · Workforce learning
Summit
2027 planned launch
A modern credentialing and exam platform for TELUS trades technicians and partner contractors — replacing two parallel in-house legacy systems.
Based on Vancouver Island, BC, Canada · working across Canada
Send the problem, not the spec.
Tell me what your team is working around today. I read every message myself and reply in writing within 48 hours.