Karina.

Product manager who builds the thing. UX discipline plus AI in the loop equals days where it used to take quarters.

Available nowToronto / remoteAI / product-led teams
About

A PM who can prototype the thing before the kickoff meeting ends.

I build with AI the same way I used to design in Figma. Sketch a flow, prototype a real working version, put it in front of a user, adjust. Days, not quarters. The bottleneck moved from “can we build it” to “is it the right thing” and that's a UX question, which is what I trained on.

The training: digital futures at OCAD and interactive media at Seneca. Game design, AR, physical computing, animation, web. I started as a freelance UI designer for startups, then took a real job at RBC designing mobile productivity apps for the people who run the bank. Six years in, I wanted to own more than the screen, so I made my way to PM.

The shipping: founding PM at a DeFi startup, pre-launch to fee-generating in four months. Now consulting as a fractional PM at an AI-for-VC due-diligence startup. Small teams, abstract problems, real shipped product.

What I'm looking for: a senior or staff PM seat at a product-led company where AI is taken seriously, shipping is taken seriously, and meetings aren't.

Recent work

AI in the loop, evals on the side.

2026 · solo~2 weeks idea-to-liveNext.js · Supabase · Gemini

mySommelier. A wine recommender that actually learns.

Five LCBO Vintages bottles you'd love this Saturday, picked by an agent that updates its taste model every time you rate a wine.

The bet.Wine recommenders fail because they don't persist taste memory. A profile that recomputes from every loved and disliked wine, plus a fresh local catalogue, beats both generic LLM chat and static review apps.

The product call. Picks first, chat as refinement layer. Conversational onboarding cut for a 5-question form plus 2 to 3 reference wines. Public profile pages with no login wall on the demo, because the product is the recommendation.

The eval work.I caught the model being confidently wrong four ways during build. Suggesting a white when I'd only logged reds. Picking a 14.5% bold when I'd said medium. Drifting above my sugar ceiling. So the system ships with three eval layers: a data-quality eval upstream of everything, a retrieval eval with four labeled personas and a hand-coded ground-truth function (NDCG@5, recall, diversity), and append-only trace logging for L2 review. Provider-agnostic, so swapping Gemini for OpenAI is a config change and the harness re-runs against the new embedding space.

What I cut. Multi-store inventory (the job is “should I buy this,” not “where is it”). Free-text NLP onboarding (kept as a bet, not an assumption). Login walls on the demo.

Let's talk.

If you're building an AI product and want a PM who can prototype alongside engineering, write the eval, and ship the thing — let's talk. Quick intro, no slides, no deck.

contact@rinalabs.ai