Fourteen years in product, most of them at the seam of federal student aid policy and the systems that run on it. Where I add the most value: the requirement is still regulation, the stakes are compliance, and someone has to decide what the software will actually claim.
Cite or don't claim. Ship the boring parts. Make it checkable. That's the whole method — applied to a law that changed loan limits for less-than-full-time students in 2026.
"...for a student enrolled on a less-than-full-time basis, the Secretary shall apply a schedule of reductions to the applicable annual loan limit, prorated by the student's enrollment intensity for the period, calculated separately for subsidized and unsubsidized amounts, net of amounts already disbursed..."
Same math, two surfaces: a full staff engine and a plain-English student estimate. Try it live ↗
The full staff engine: SOR %, per-term Sub/Unsub/Grad PLUS, disbursement schedules — plus a plain-English student estimate on the same tested logic.
Open the calculator →A COD annual-update portfolio hub: schema diffs, report-change dashboards, field logic, and the product narrative behind each year's release.
View the hub →An advanced student SOR estimate with scenario comparison, plus a lifecycle tracker for how a reduction evolves across a student's enrollment. Both run on the same engine, with no forked logic.
Open the advanced estimate →Ask a financial aid regulation question and get the official citation, an excerpt, and a link to the source. Refuses student-specific questions.
Ask a question →I'm building this in public, one layer at a time. Citations come first; the model comes last and has to quote its sources.
Ask a question, get the official regulation and a link. No AI model, $0 per question, tested against a golden set.
FSA Handbook, Dear Colleague letters, and guidance, searchable by meaning, not just keywords.
A plain-English answer that must quote its citations, and refuses when the sources don't support one.
The same five moves, every release. Nothing ships that skips a step.
Every step is source-labelled and regression-tested against published fixtures. Methodology & sources ↗
An open calculator for the OBBBA less-than-full-time Direct Loan reduction, built for financial aid offices and the students they serve.
Turning the yearly Common Origination and Disbursement technical update into readable, gated, brand-owned segments.
An MCP server and versioned public contract so AI agents can answer Schedule of Reductions questions with the real engine instead of guessing.
Every number this site produces traces back to a labelled source, and where the regulation is unsettled the tool says so instead of guessing quietly.
Students and financial aid staff need opposite levels of detail from identical math. One tested engine, different surfaces, no forked logic.
Public scenarios, an OpenAPI contract, a scenario-challenge process. If I am wrong, I want it to be cheap for someone to prove it.
Versioning, disclaimers, release markers, and regression fixtures are the product when the domain is compliance.