Ben Siverly

Property-management software for 20k+ residential operators.

• Rebuilt the rental application end to end. Submission rates went from 44% to 70% with a zero change-failure rate, and properties on the new flow fill vacancies 3.4 days faster than offline workflows.
• Shipped follow-up for incomplete applications inside the authenticated app, not over email or text, so every interaction stays auditable for screening compliance. Customer satisfaction scores on rental applications rose 18% after it shipped.
• Built an API for securely moving applicant PII between platforms to pre-fill rental applications. We launched with Zillow as the first use case and got to 70% customer adoption. Pushed to wind it down when the frictionless path buried property managers in unqualified applicants.
• Sole PM for the application line, ~$40M in payments a year: strategy and roadmap across the applicant and property-manager experiences, three stakeholder types, and two codebases that have to move together. The payer and the user are different people, and the product has to serve both.
• Built a verified jobs-to-be-done map from 50–100 customer calls a year: a structured model of what renters and property managers actually need. Every product bet traces back to it. The transcript is the easy part. The model is the work.
• Building a spec-driven way of working with my team: we write down how the product should behave, pressure-test it before any code, and gate every release behind tests and evals.

Modern HSA platform for tax-advantaged health spending.

• Built the partner APIs and migration pipelines that moved $50M+ in HSA assets across financial institutions with zero data loss, cutting the work from two migration teams down to one.
• Built the platform under it: money movement across banks and card networks, processor integrations, ledgers, and the data model partners built on. That let them offer compliant HSA accounts without building the infrastructure themselves.
• Stood up the company’s first data-science function and data-services platform; promoted to PM in 2021 as the work became the roadmap.

Enterprise software; consumer-data research in Data Cloud.

• Rebuilt the model-validation methodology on a representative U.S. population sample, correcting a data skew that had been distorting how we measured model performance.
• Automated the ad-viewability data pipeline with Python and the JIRA API, replacing extraction and cleaning that had taken two analysts full-time.