Alexander Laicovsky

Internal dashboard an advertising client's team uses daily. The hard part was turning several mismatched data sources into one view they could actually trust in real time.

Data comes from multiple platform APIs and Google Drive, each with its own schema and OAuth token lifecycle to keep alive. I normalised it in a Node.js backend and made the daily cron idempotent, so a failed run can retry without duplicating records — the difference between a dashboard the team trusts and one they quietly stop using.

AI movie recommendation app. I built the recommendation engine and the cross-platform mobile app, using OpenAI to turn a few swipes into suggestions that actually fit each user's taste.

I chose a C# backend for the recommendation logic — based on the user's history, it builds a prompt for the OpenAI API. The React Native app uses a tinder-style swipe to capture taste signals fast, then sends them to the engine, with one consistent experience across iOS and Android — the process runs in parallel, so the user never stops swiping and receiving new recommendations.

Fast, maintainable static web presence for international clients. Built with Astro.js (like this portfolio).