~/playground
Hi, my name is Noemi Rugar
Senior PM × AI builder
> vibecoding side projects…
This site is my solo playground — AI experiments and vibecoding projects I build on my own. My day job (and stealth build) is leading proper product teams.
01
Live in the lab
Solo vibecoding projects
AI
Playground builds
Just me + Claude · Cursor
10+
Yrs leading teams
Internal & external squads
Solo vibecoding experiments — just me, AI, and too much coffee. Each lives in its own repo.
AI ad planner & diagnostician for small businesses
Meta · TikTok · Google split
Know what to automate — and what to leave alone
scanning workflow...
Side quests, whims, and proofs of concept
$ npm run side-quest
→idea.sh◌
→sketch.tsx◌
→vibe.test◌
→ship???◌
→refactor.coffee◌
const nextProduct = await ship({
status: "stealth",
team: "dev squad + PM",
mode: "proper build",
launch: "soon™"
});
A stealth product I'm building with my dev team. Not vibecoded. Not a side quest. A proper product, shipping the way teams actually ship.
I've spent a decade leading digital transformation across MEA — marketplaces, loyalty, eCommerce at scale. I work with squads, vendors, and cross-functional stakeholders to ship real products the traditional way: roadmaps, backlogs, sprints, and teams that actually build. This lab is my personal space — I vibecode side projects here on my own, with AI as a force multiplier. My stealth product is different: I'm building it with a dev team. I own the strategy and product; they own the engineering. Two modes, same obsession with shipping.
The domain changes. The method doesn't. I've built loyalty platforms across 12 markets, D2C storefronts, and on-demand marketplaces — and the approach was the same each time. Knowing how to find the answer travels better than already knowing it.
01
I start with two questions before anything else: what problem are we solving, and what does success look like? The feature is the last decision, not the first. Getting the problem and the expected outcome agreed up front is what keeps a project from drifting.
02
Once the problem is framed, I sit with everyone in the workflow — the people who actually do the work, day to day — to understand what they do and how they do it. Every stakeholder has a need, and a solution built without hearing all of them breaks the moment it meets reality. Only then do I lock in the confirmed problem and the solution, with the people who have to live with it.
03
Not every good idea earns its place next. I weigh each piece on two axes: what it returns commercially, and what it costs in time and effort. Something expensive and slow that doesn't move revenue gets sequenced below something simple, or something with a clear revenue outcome. Prioritization is where strategy becomes a real plan.
04
I want a working prototype in front of people as soon as possible. Feedback on something real beats months of feedback on a document, and it means we can pivot while pivoting is still cheap. Building the wrong thing slowly is the most expensive mistake a team can make.
I use AI to move fast — research synthesis, first drafts, prototyping, code. But the decisions that matter stay with me: what to build, who it's for, what to cut, and which tradeoff is worth accepting. AI compresses the distance between an idea and a working version of it.
Hand me a domain I've never worked in and I'll be useful in a week.
What I bring to the table — from enterprise transformation to the tools I use with real teams.
Where I've led internal teams and external partners to ship at scale — click a role for the full story.
MEA
Europe