Onsense AI builds dedicated ML

Onsense AI builds dedicated ML infrastructure for equipment repair — private LLMs and AI agents trained specifically to help technicians, OEMs, and support teams diagnose and resolve repair issues faster. Beyond the core web and mobile products, the platform needed to move from AI as a bolt-on feature to AI agents as a core part of how support, diagnostics, and technician workflows actually run.

What I work on

I work across the full stack — architecting the platform's Turborepo monorepo, building Next.js web dashboards and NestJS backend APIs, and shipping the React Native (Expo) mobile app for iOS and Android. Alongside that, I'm part of a small team (2-3 engineers) building out the Backend for ERP system, developing frontend for apps and building mobile apps for android and iOS

Stack:

Next.js and NestJS for web and backend

React Native (Expo) for mobile

AI integrations

PostgreSQL, Redis, Docker across the platform

GitLab CI/CD and Playwright/Jest for testing and deployment

What shipped

A production platform where AI agents aren't a separate feature but a working part of the support and diagnostics experience — backed by a full-stack team handling everything from infrastructure to the agent layer itself.