Making complicated systems work at scale.

I work on the systems behind sales compensation — where business policy, data, engineering, finance, and the experience of the person getting paid all have to agree.

A large part of my work has been taking a new compensation model from a 500-person pilot to ~23,000 sellers: translating evolving business policy into product requirements, navigating architecture and data dependencies, and carrying the product through build, launch, and hypercare.

From a 500-person pilot to ~23,000 sellers.

What I’ve found most interesting is everything that happens around the edges of the system — the ambiguous policy decision, the dependency nobody owns, the number that doesn’t reconcile, or the production edge case that forces you to understand how the entire thing really works.

More recently, I’ve been exploring how AI can make these systems easier to navigate: from conversational compensation support and smarter escalation to traceability experiences that help someone understand why a particular outcome happened without manually piecing it together across systems.