MatterWorx

MatterWorx is a vendor management platform for staffing. The agencies, shifts, credentials, timesheets, and invoicing that used to live in separate tools now run in one automated system.

The problem

Running a staffing operation means juggling disconnected tools: one for onboarding agencies, another for scheduling shifts, a spreadsheet for credentials, something else again for timesheets and invoicing. Nothing reconciles on its own, and the manual handoffs between those tools are where errors and delays creep in. The platform needed to pull all of it into a single system and automate the workflows that tie those pieces together.

What I built

I work across several domains of the platform, on both the operational workflows and the interfaces that drive them:

• End-to-end operational workflows covering agency onboarding, shift scheduling, credential tracking, invoicing, and compliance, automated so the steps that used to be manual now hand off on their own.
• Geofencing and shift management, so on-site shifts are tracked against real locations rather than on trust alone.
• Timesheets and candidate management, wired into the same scheduling and invoicing flows so the data stays consistent from start to finish.
• One consolidated platform that replaces the patchwork of separate tools a staffing operation used to run on.

How I built it

MatterWorx is built AI-natively. Claude Code and Cursor are the primary build engine, and close to 97% of the codebase is AI-generated. That pace only holds up because the workflow is built for it. Domain-driven design keeps the boundaries clear, and test-driven development keeps the AI-generated output honest, so what ships is production-grade rather than a fast first draft. It let one engineer deliver at the pace a full team would normally need.

Stack

Next.js and TypeScript on the front end, .NET on the back end, deployed on AWS with Docker. Domain-driven design and test-driven development throughout.

Impact

Tooling that used to be spread across several disconnected systems now runs as one platform, with the workflows between onboarding and invoicing automated rather than stitched together by hand. Holding the AI-native build to production standards through DDD and TDD is what made a project this broad possible at solo pace.