Use case · Finance and accounting

A reconciliation that ties out can still be wrong.

Finance agents pass automated checks and still book the wrong number: the right amount in the wrong period, an exception cleared instead of explained. Lasker sources the signal that catches it: evals by credentialed accountants, traces of how the close is really done and licensed, de-identified workpapers.

For teams training or evaluating agents for the close, reconciliation and audit.

Credentialed reviewersCPA, ACCA, CA, CIA
Rubric per taskWritten before scoring
Rights record per datasetTitle, consent, personal data
Pilot before you commitOne process, one sample
// Where models break

Where finance agents get sent back.

Tying out a balance is the easy part. These are the places a controller or an audit senior returns the work.

Reconciliation

Matching bank lines to the ledger is easy. Timing differences, partial payments, fees netted from receipts and FX each need a judgment and a note.

Exception 14: cleared, not explained.

Cut-off and accruals

An expense has to land in the period the service was received, with support. The right amount in the wrong month is still an error.

Invoice dated May 2 for April service: accrue in April.

Journal entries

Every entry needs the right accounts, support and an approval trail. A plausible entry with no support fails review.

JE-311: no support attached.

Revenue and leases

ASC 606 and IFRS 15 for revenue, ASC 842 and IFRS 16 for leases turn contract terms into numbers. The answer is in the contract, not the invoice.

Contract 22: performance obligations not separated.

Audit sampling

Choosing a sample, testing it and documenting the result so another auditor can follow it.

Sample size not tied to the risk assessment.

Materiality and explanation

Knowing which variance matters and writing a reason a reviewer accepts.

Rent expense up 18% month on month: no explanation.

Public data has the financial statements. It does not have the reconciliations, workpapers and review notes behind them.

Automated checks confirm that numbers tie out. A controller asks whether they are right, in the right period and supported.

Workpapers belong to the companies that prepared them and hold client and personal data, so they have to be licensed and cleaned, not scraped.

// What a production agent needs

Six needs, and where the data comes from.

For teams whose agents do close and audit work that someone has to sign off.

Know what a reviewer rejects

Review notes on real work: what was flagged, why, and how it was fixed.

Source: review notes from controllers and audit seniors

Follow the real process

Step-by-step traces of a close task, with the reasoning at each step.

Source: workflow traces by credentialed accountants

Handle the messy cases

Exceptions, partial matches and unusual entries, not clean textbook data.

Source: licensed reconciliations with exception notes

Cite the standard

Answers tied to the US GAAP or IFRS rule they rely on.

Source: expert evals with the standard cited

Pick the better answer

Two outputs for one task, ranked by an accountant, with the reason.

Source: preference data from reviewers

Work across ledgers

Different charts of accounts, ERPs and close calendars.

Source: records licensed from companies of different sizes

// What we deliver

Six datasets, scoped per pilot.

Expert-made data is open for pilots now. Licensed company records are cleaned and come with a rights record. We deliver in your schema.

DatasetWhat is in a recordUsed forFormat
Expert evalsAgent output scored against a written rubric by a credentialed reviewerEvals, release gatesScores + rationale, JSONL
Workflow tracesA close task done step by step, with the reasoning at each stepDemonstrations, fine-tuningJSONL + source files
Preference dataTwo outputs for one task, ranked, with the reasonReward modelsJSONL
Review notesReviewer comment on a workpaper, the fix and the final versionVerifiers, critique modelsJSONL + workpapers
Licensed reconciliationsDe-identified reconciliations with each exception and how it was resolvedTraining, evalsCSV + JSONL
Close proceduresClose calendars, checklists and procedures as a company runs themPlanning, groundingPDF + extracted steps
// Sample record

What records look like.

An expert eval of an agent’s bank reconciliation, and a licensed reconciliation with its exceptions.

  • scoresPer rubric criterion, so you can train on what failed
  • verdictPass, corrections or reject, as a reviewer would sign
  • exceptionsEach difference with its type and how it was resolved
  • frameworkUS GAAP or IFRS, recorded on every record
  • rightsSource, owner and license scope, per record

Illustrative. Field names are our draft schema, and we adapt it to your pipeline in the pilot.

ev-0412.jsonIllustrative
{
  "record_id": "ev-0412",
  "task": "Reconcile operating cash to GL, April close",
  "framework": "US GAAP",
  "reviewer": {
    "credential": "CPA",
    "verified": true
  },
  "scores": {
    "accuracy": 4,
    "cut_off": 2,
    "exceptions_explained": 2,
    "support": 3
  },
  "finding": "Ties out, but a May 2 deposit clears an April gap.",
  "verdict": "corrections",
  "rights": {
    "source": "expert-authored",
    "personal_data_removed": true
  }
}
rc-1187.jsonIllustrative
{
  "record_id": "rc-1187",
  "owner": "company-07",
  "period": "2026-04",
  "account": "operating cash",
  "framework": "IFRS",
  "exceptions": [
    {
      "type": "timing",
      "amount": 4210.00,
      "resolution": "deposit in transit, cleared May 2"
    },
    {
      "type": "fee_netted",
      "amount": 35.50,
      "resolution": "bank fee booked to expense"
    }
  ],
  "rights": {
    "license": "non-exclusive, training + eval",
    "personal_data_removed": true
  }
}
// Expert review

How reviewers score finance work.

The questions a controller or an audit senior asks, written down before scoring so two reviewers can agree.

DimensionWhat the reviewer checksEvidence per finding
AccuracyAmounts, accounts and signs are rightLine reference
Period and cut-offEntries land in the right periodDate evidence
Standard appliedThe US GAAP or IFRS rule fits the factsStandard cited
ExceptionsEvery difference is explained, not just clearedNote per item
SupportA reviewer could sign it offMissing support listed

Reviewers hold the credential the task needs, checked with the body that issued it. Items with disagreement go to a second reviewer and are adjudicated.

// Pilot

A pilot on one close process.

Pick the process where your agent is weakest. The pilot answers whether this data moves it.

Scope

One process, such as bank reconciliation or accruals, and the framework: US GAAP or IFRS.

Week one

Sample

A small set of evals and traces, scored against a rubric you approve.

You test it in your pipeline

Decide

Full set, ongoing work or stop. Price is set per dataset after the sample.

No commitment before

Expert evals and traces are open for pilots now. No company dataset has been licensed yet, and we will not quote volumes until owners have signed.
// Inside your walls

When the records cannot leave.

Banks and regulated companies often cannot send customer or financial records outside. Our reviewers work inside your environment, on your devices or virtual desktops, and only the findings you approve leave. It suits the close, reconciliations and credit work, scored by the same credentialed reviewers.

// FAQ

Questions labs ask.

Data handling and our commitments are in the Trust center.

Who are the reviewers?

CPAs, ACCA members, chartered accountants and internal auditors, matched to the task and checked with the body that issued the credential.

US GAAP or IFRS?

Both. The framework is set per task and recorded on every record.

Can we bring our own tasks?

Yes. Send the tasks and your agent’s outputs, and reviewers score them against a rubric we agree with you first.

Where do licensed records come from?

From the companies that own them, under a non-exclusive license with a rights record. Client material is excluded unless the client agreed in writing, and personal data is removed before release.

Can reviewers work inside our company or bank?

Yes. Reviewers work in your environment on your systems, under your access controls. Records stay with you, and only the findings you approve leave.

Can a model trained on it ship commercially?

The license scope is written per dataset: training, evaluation or both, and whether a model trained on it can be used commercially.

How is it priced?

Per dataset or per batch of tasks, after the sample. We do not publish ranges until we have closed deals to base them on.

Keep finance records? Your company keeps ownership. License close and reconciliation records on your terms.

License your records →
// Scope a pilot

Tell us where your agent fails.

Send the process, the framework and the volume. We reply with a rubric draft and what a pilot would include.