The work
Nine projects. Each started as a real job for a real business — sometimes my own. Click any one for the full story, including what's working, what isn't yet, and what it would look like to do something similar for you.
A month of work, three hours of work
Over 6,000 SKUs of product data migrated between two inventory systems in about three hours.
Read the story →
Eighteen years of email, turned into a pattern library
Over 500,000 messages in my email archive, filtered down to 736 high-signal customer recommendations — now a pattern library my sales and marketing partner's AI assistant queries live, every week.
Read the story →
A CV anonymizer that strips identity before any AI reads it
Raw resumes stop at a Montreal server. The cloud AI only ever sees the scrubbed skills and experience, and the candidate's privacy never leaves Canadian soil.
Read the story →
A training module, shortened on demand
Two production runs in the first month — MacKay CEO Forums' CHRO session and Alberta Health Services' APL team. Same cutting protocol, two very different pipelines.
Read the story →
A seven-stage protocol for bilingual website redesigns
Seven stages my AI pipeline runs alongside me — deep research, content triage, new foundation, bilingual build, three-phase QA. A fourteen-page bilingual replacement, end-to-end, in a couple of hours.
Read the story →
An anonymizer app that runs on your laptop
A 60-minute interview transcript takes a research assistant 2 to 4 hours to anonymize by hand. This tool does it in about 5 seconds — and the same edge AI scrubs any document with PII your team can't send to the cloud.
Read the story →
An agent that preps the room before you walk in
I run AI enablement workshops, and I always wanted to know what the people in the room needed from AI before I walked in. So I built an agent the participants email before the workshop — it collects what they want AI to do, explains what AI can and can't do for their task, coaches them through getting it working, and hands me a briefing built out of real, named pain points the participants already own.
Read the story →
A window-and-door takeoff pipeline for a Quebec fenestration installer
Architect's PDF in, priced takeoff out. The hard part was building a visual recognition pipeline that produces the same answer for the same drawing, every time — because a takeoff with forty-seven windows must be exactly forty-seven windows, not forty-six or forty-eight.
Read the story →
Which hundred SKUs to fix first — and what to do with each
When a catalog business tells me "we have about a hundred bad SKUs but we don't know why" — this is the protocol that gets to a defensible action plan in five to seven hours. It catches the analysis traps that make the data lie, and only ships recommendations with the evidence to back them up.
Read the story →Not sure which one is closest to your situation?
Most engagements start with a 20-minute call where I listen and you describe one real thing you'd like to get done faster.