subinc is billing infrastructure for multi-tenant SaaS: the correct, gateway-agnostic core that most teams rebuild badly, built once, properly.
Not in the happy path, in the retries, the webhook replays, the partial outages, the currency edges, the test rows that leak into production. That is where money goes missing and audits fail.
subinc is built for those failure modes first. Multi-tenant isolation at the database, idempotency on every mutation, a double-entry ledger in minor units, and a hard test/live boundary that batch jobs respect.
The payment edge sits behind adapters, so the processor is a config value, not a re-platform. You keep leverage; the books stay correct.
The ledger is the demo. If the books drift, nothing else matters. Every feature defends the invariant that debits equal credits.
We never couple a team to one processor. The payment edge is adapters, so you keep leverage and we keep our promise.
We show the mechanism doing its job, not adjectives about it. Demonstration data is labeled; commercial claims are never invented.
The buyer reads the API before the pitch. So the API, the docs, and the failure modes come before the marketing.