For accounting firms and CAS teams

Your name is on the number. The system has to show its work.

Your people have already started using AI, and some of them have not mentioned it. Meanwhile the carrier questionnaire asks, §7216 and the Safeguards Rule apply whether or not anyone has written the policy, and the first annual evaluation of the firm’s quality management system falls due by 15 December 2026 — which asks you to describe how you monitor what your people actually do.


What it does

The modules that matter when the entities are clients

Each one runs a real workflow, in the order it runs. Every one of them ends with a person, because that is the part your licence cares about.

Transaction classification that learnspredict queue for review approve or correct learn from the correction Coding predictions queued for an accountant’s approval, with every correction captured and fed back into the categoriser that made it.

Your firm’s coding judgment accumulates as policy rather than staying in one person’s head.

Bank reconciliationimport match confirm or unconfirm compute tie-out complete Reconciliation as a session with a lifecycle: proposed matches a person confirms, a computed tie-out, and a completion that only passes when it balances.

Reconciling items carry a history and an owner instead of living in a workbook copied forward.

Accounting schedulesdetect on a transaction propose human approve generate entries recompute on override Prepaid and fixed-asset schedules detected from transactions, confirmed by a person before anything is recognised, with entries regenerated whenever a term, life or classification is overridden.

The schedules that turn into audit adjustments stay current instead of being rebuilt each year end.

Leases and revenue — ASC 842 and ASC 606draft measure gated commencement entry amortise tie out to the ledger Terms editable in draft; activation measures and persists the schedule, and the commencement entry is always gated for approval. Disclosures and tie-outs come out of the same records. Revenue modifications re-allocate without touching posted periods and surface the catch-up explicitly.

A contract change shows you its catch-up instead of quietly restating a month you already closed.

A documented closegenerate the period assign verify against live data complete hard lock A checklist generated for each period with dependencies and due dates, assigned to named preparers, auto-verified against the live ledger, and locked once the period closes.

The process is written down, not living in one person’s head over one busy season.

Preparer and approver, separatedper-entity review policy role floors materiality overrides named approval Review policy set per client, with role floors and per-check materiality overrides, and separate preparer and approver roles wherever the control requires segregation.

“Who approved this entry, and on what basis” has an answer before anyone asks it.

Your written AI policy, and the register under itinventory the tools exposure, terms and books findings the policy Every tool touching client work, what it is exposed to, whether its terms have been read, and how it sits in the books — with findings that cite 16 CFR §314.4(f) and AICPA ET §1.700.040. The Register assembles in your browser and the answers never leave it.

The questionnaire stops being a scramble, because the inventory already exists.

Or build it yourself — The Sprintdownload answer eleven business questions hand the plan to your coding agent The platform source itself rather than a description of it: the tool registry and its integrity gate, the ASC 606 and ASC 842 engines, the four runtime controls, and a guided builder that turns your decisions into build instructions. One firm, unlimited seats.

If your firm would rather own the thing than rent it, this is the foundation to start from. See what is inside →

Modules you do not need are switched off rather than shipped as clutter. What each one does across your client base — their entities, calendars, charts of accounts, materiality thresholds and your review steps — is configured during the build.


The rule, before the modules

One rule holds across every workflow above

The AI retrieves, analyses, drafts and proposes. Deterministic code calculates and checks. An authorised person approves anything that reaches the ledger. There is no unsupervised posting, and no setting that turns it on.

Four-tier action gate

Every action is classified into exactly one side-effect tier. Ledger and customer-facing writes are proposal-first, human approval only. An unrecognised action fails closed into the most-gated tier.

Preparer and approver

Review policy set per entity, with role floors and per-check materiality overrides, and separate preparer and approver roles wherever the control requires segregation.

Readable proposals

Each proposal carries before and after, its source data, risk level, the role required to approve it, a content fingerprint, an expiry, and its execution record.

One audit chain

The prompt, the run, each tool call, the proposal, the approval or rejection, the sync — recorded as one chain you can walk backwards.

Circuit breaker

When a task’s accuracy degrades, the system revokes that task’s autonomy and alerts. It can only demote. It can never promote itself.

Grounded answers

Answers cite period, basis, source reference and sync timestamp. Prose claims are verified against facts captured at the tool boundary; anything unverified is marked.


How firms use us

Two ways in. Both end with you owning it.

The rungs are an engagement — we run the client work while your people watch it happen, and at twelve months they take it over. The shelf underneath is the same thinking sold as materials, for a firm that would rather build it now. Both end in the same place.

01
Free·90 seconds·nothing leaves your browser

The AI Register

Name the AI tools your people already use on client work and the register builds itself — the inventory the first annual evaluation of your quality management system asks for, due 15 December 2026.

You leave withA workpaper listing every tool, its terms, and how it is treated in the books.
02
Free·45 minutes·five a month

The Close Teardown

Bring the client close that costs you the most hours and we go through it together. No demo, no pitch, and no obligation at the end of it.

You leave withOne written page about that close, sent afterwards.
03
$7,500·one close cycle

The Pilot

Pick the client close that costs you the most every month, and we run it — not a report about running it, the close itself, on your deadline, against their books. You watch it happen on an engagement you already own.

You leave withThat month’s close delivered on time, the evidence pack behind it, and a written read on which of your other engagements it would fit.
04
$5,000 a month·three months to start

The Run

We keep running it, then the second engagement, then the third. An evidence pack every close — what ran, what it proposed, who approved it, what changed. Your review steps, your materiality, your sign‑off.

You leave withClient work that arrives on the date it is due, and a written record of every judgment inside it.
05
At twelve months

The Handover

After a year of it running, your own people take it in‑house — the source, the training, and support while they settle into it. This is the plan from the first day, not the thing the contract is written to avoid.

You leave withThe source, on your own infrastructure and your own model keys. Nothing to renew, and no login to lose.

If you would rather own it outright from the start, we build it that way instead — scoped against what your practice actually runs, so it carries no published price. Most firms do not start there. The teardown is where you find out whether you should.

Or build it yourself

The same ideas, sold as materials rather than as an engagement. Not a decision you have to make now — the teardown costs nothing and is the fastest way to tell which one fits.

$25·one evening

The 101

What the tools actually do, what they cannot do, and the vocabulary the rest of this is built on. Written for someone who has never opened a terminal.

You leave withEnough to judge a vendor’s claim, which is most of the value.
One payment·yours permanently

The Sprint

You answer eleven questions about how your firm actually works — which system holds the books, who may approve what, whether anything may reach a client ledger at all — and it writes the build instructions for your own coding agent. Nothing gets built until you approve it.

You leave withA perpetual licence to the source — one firm, unlimited seats, on your own infrastructure. Not a subscription to ours.
$5,000 per firm·four live weeks·capped seats

The Bootcamp

The Sprint, run with you, over four weeks — the same build, with the judgment calls made out loud and your questions answered while you are inside the file rather than after.

You leave withThe software, and people in your firm who can change it without calling anyone.

Both paths end in the same place: your firm owning what it runs on.


Where to start

Find the layer that is least ready to carry it.

Twelve questions across six control layers. About six minutes, and the result stays on your screen — sharing it is optional.