About ARC

ARC separates what a contract says from what the accounting requires, and assigns each part of the work to the party best suited to it.

AI interprets contract evidence. When a contract PDF is provided, AI helps identify the relevant contractual facts — parties, dates, promises, pricing and billing terms — and points to the passages that support them.

Deterministic accounting engines calculate results. Allocation, revenue schedules, contract balances and journal entries are computed by fixed application logic, not generated by the model. The same inputs always produce the same outputs.

The accountant owns the judgment. Every AI-drafted fact or conclusion remains reviewable and editable. Open review items require an explicit decision by the accountant before an analysis can be finalized.

Revenue recognition depends on judgment, and judgment is only as reliable as the evidence behind it. ARC is built around that idea: conclusions stay tied to their source evidence, calculations are validated and reconcile across workpapers, and reviewable items are resolved explicitly rather than accepted silently.

Where source evidence is insufficient, ARC surfaces the gap for review rather than silently treating an assumption as established fact. Automation assists the accountant, but control over the analysis stays with the accountant.

ARC was built as a portfolio project to demonstrate how technical accounting and software can work together. It brings together ASC 606 application, product and workflow design, AI-assisted contract review, testing and validation of accounting logic, and the traceability expected in reviewable accounting workpapers.

ARC is a portfolio and workpaper application. It is not a commercial accounting system and is not a substitute for professional accounting judgment. ARC prepares journal-entry workpapers, but it does not post entries to a general ledger, issue customer invoices or produce financial statements.