At Revotas, a B2B marketing and customer-experience SaaS, I built the admin platform, the on-site search, the recommendation rendering engine, a no-code widget builder and the mobile push client, plus the rollout tool that took search to 271 brands, an email builder block and three years of work in the shared panel. The code and client data belong to the company, so each page recreates the product as an interactive demo with synthetic data.
A back-office console written from scratch, with cited AI answers over internal documents
A back-office console I wrote from scratch as its only author. Staff run accounts, contracts, invoices, support and billing from one place.
One script that puts search and autocomplete on any storefront, live on 271 brands
A search overlay and autocomplete that attaches to a storefront's own search box with one script, configured per brand without code.
A CLI that took search from one brand to 271 without touching a storefront
A TypeScript CLI that installs Search V2 through the loader each site already runs, and dry-runs by default.
Recommendations rendered in the store's own card design and carousel
One config-driven script that copies a real product card from the page, fills it with recommendation data and starts the store's own slider on it.
A no-code builder that compiles a marketer's settings into a storefront script
Deal of the Day V2, built from scratch: a builder that turns settings into a self-contained storefront script, with a live preview.
Paste a product URL and the email block fills itself in
A block for the panel's drag-and-drop email builder that fills in a product's title, price, image and description from its URL.
A guard that stops saves landing on the wrong brand, and three years of panel work
Work inside the company's main marketing panel, a React codebase with years of history and many contributors.
Push notifications that open the right screen from any app state
The React Native side of push for the platform's mobile clients, from token registration to deep-link routing.
Most of this work runs on other companies' websites, so it has to fit hundreds of storefronts I don't control.
Search, recommendations and widgets each ship as one script. Per-brand behavior lives in configuration: selectors, layouts, synonyms, strategies, labels. A new brand is a config file, not a code fork, and teammates outside engineering can set one up from the docs.
Because the scripts load on third-party pages, the same problems come back on every project: the host's CSS leaking in, consent banners and other vendors' scripts racing ours, slow networks, Turkish text and number formats, and keyboard users. The case studies show how each system handles them.
Every company, person, product, price and chart value in the demos is invented. Nothing calls a real API. The numbers in the page headers (line counts, commits, the 271 brands) come from the real repositories and rollout records.
I built these systems; I did not build everything around them. Each page marks what I owned and what I integrated with, for example the recommendation strategies and the push backend, which come from other teams.
Open to full-stack and applied AI roles, remote or hybrid.
Tell me about the role, the team and the problem.