OnboardOS — Learn the Job by Doing It

OnboardOS is an AWS-powered workplace training simulation platform that turns passive employee onboarding into hands-on experience.

Traditional onboarding often relies on documentation, videos, and shadowing. New hires may understand the theory without having practiced the decisions and workflows they will encounter on the job.

OnboardOS gives employees realistic, controlled workplace assignments where they investigate problems, interact with simulated systems, make decisions, and receive measurable feedback.

What we built

Our first executable simulation puts an employee in the role of a cloud engineer investigating a checkout API timeout incident.

The employee can inspect incident evidence, read operational runbooks, examine diagnostics, modify a simulated service configuration, and execute automated tests. Their work is recorded and submitted for manager review.

Instead of treating an AI-generated explanation as proof of competency, OnboardOS uses deterministic evaluation to check the employee's simulated work.

Technology

OnboardOS uses Amazon Cognito for authentication, API Gateway and AWS Lambda for backend operations, DynamoDB for persistent application state, and a separate Lambda-based scenario evaluator.

Amazon Bedrock powers the AI coworker architecture, which is being developed to provide contextual workplace guidance using scenario evidence and configurable coaching preferences.

The frontend uses React and TypeScript with Vercel hosting.

Business potential

Our initial customers are companies onboarding technical employees. OnboardOS could help reduce repetitive mentoring, standardize training experiences, and provide managers with evidence of task readiness.

The long-term direction is a library of reusable, employer-configurable workplace simulations across multiple professional roles, with organization licensing and usage-based cloud infrastructure costs.

Current scope: One executable cloud-engineering simulation forms the working prototype. Broader scenario authoring and the newest manager-configurable AI coaching features remain in development.