Work with me

I've shipped an AI product on the Claude API — content moderation, prompt architecture, the whole stack. I also research how AI safety systems behave versus how they're documented. If you're building on LLMs and could use someone who's already been through the hard parts, I'm available for consulting.

What I do

AI Red-Teaming & Safety Evaluation

Adversarial testing of AI systems. I find where your content moderation breaks, where your safety guardrails have gaps, and where the model does something different from what your documentation says. Methodology comes from offensive security testing applied to AI content systems, and I've published the research.

Prompt Architecture & Engineering

SkeinScribe runs on a prompt architecture that handles branching narratives, entity tracking, content rating tiers, and multi-layer moderation — all on the Claude API. If you need prompt systems that work reliably at scale, not just in a demo, that's what I build.

AI Content Moderation Design

Building content safety systems and testing where they break. I've done both sides, which turns out to be useful.

AI Product Consulting

I've taken an AI product from idea to paying customers. The hard part wasn't the API integration — it was everything else. Architecture decisions, cost optimization, content pipelines, and the hundred things that aren't in the documentation. I can probably save you some of that time.

How I work

Remote and async, based in Ottawa (ET). I'll tell you something won't work before you spend money finding out. Available for:

• Advisory / fractional — ongoing second opinion for teams making AI product decisions without wanting another full-time person.
• Project-based — scoped work with a deliverable. Prompt architecture, moderation system design, that kind of thing.
• Red-team engagements — I try to break your AI systems and write up what I find. Time-boxed, report included.

Let's talk →