creditboard
The Tool · CRAA
Independent loan review, for community institutions

Make your next exam boring.

CRAA gives a community bank or credit union an always-on, independent loan-review function — consistent risk ratings and examiner-ready documentation on every credit — for less than the cost of a single outsourced engagement. A qualified human always owns the final call.

Pass → LossUCS classification framework
OCC · FDIC · FRBShared interagency grounding
100%Reviews logged to an immutable trail

The line we do not cross

AI can produce the work. Only a human can own the conclusion.


The problem

The problem every credit officer already has

Required

You must have independent loan review

The 2020 Interagency Guidance on Credit Risk Review Systems requires a qualified, independent review function. A small institution cannot realistically staff a bench of seasoned reviewers.

Expensive

Outsourced review runs $25k–$60k a year

And it arrives once or twice a year as a static report — a snapshot that is stale the week after it lands.

Inconsistent

Ratings live in one person’s head

Grades vary by which officer originated the loan. When that person retires, the judgment built over a career walks out the door.

Exam risk

Stale, undocumented ratings draw findings

Inaccurate or untimely risk ratings are the most common way a community institution gets criticized in an exam — and they flow straight into the CECL allowance.


How it works

Deterministic where it must be. Judgment where it helps. Human where it counts.

CRAA is not one black box. It is three layers, each doing what it is genuinely best at — which is what makes the output defensible to a model-risk team and an examiner.

1

Deterministic rules engine

Code · reproducible

Ratios — DSCR, leverage, coverage, LTV — are computed from the raw financials, not trusted from free text. Hard policy gates are enforced in code, not by persuasion: a DSCR below 1.0 cannot be Pass; an unwaived covenant breach cannot be Pass. Same inputs, same result, every time.

2

AI judgment layer

Grounded

Within those guardrails, the model performs the qualitative work that is inconsistent and un-staffable today: the five-Cs synthesis, red-flag detection, accrual-status reasoning, and a draft narrative in examiner style — each assertion tied back to a specific data point and a cited regulatory passage.

3

Human ownership

Accountable

The reviewer accepts, or overrides with a documented rationale. Nothing is a rating until a qualified person makes it one. Every decision — AI recommendation, human decision, override reason — is stamped and written to an immutable audit trail.

PassSpecial MentionSubstandardDoubtfulLoss

What it changes

What it changes, by who’s asking

Chief Credit Officer

An answer for every credit, in seconds

When the examiner asks “why is this still a Pass?”, the evidence, the cited classification, the five-Cs rationale and the override note are already there — consistent across the whole book.

Loan Reviewer

Cover a bigger book, skip the grind

Financials are spread for you and the first-draft narrative is written. You spend your time on judgment and exceptions — and a junior produces senior-quality drafts.

Board · Audit

Governance you can point to

An immutable trail, a rating-migration view of the portfolio, and a validation record behind the tool itself — the artifacts that make an exam a formality.


Regulatory grounding

Grounded in the framework your regulator already uses

The core classification — Pass, Special Mention, Substandard, Doubtful, Loss — is the shared interagency standard used by the OCC, FDIC and Federal Reserve, drawn from the OCC Comptroller’s Handbook. The regulatory layer is a per-institution profile, so grounding matches your charter.

Banks — live

OCC · FDIC · FRB

UCS classifications, five-Cs framework, accrual-status determination and examiner-defense notes, grounded in embedded interagency guidance.

Credit unions — in development

NCUA profile

Aligned to NCUA supervisory expectations where they diverge from the interagency banking framework — validated before we claim coverage, never assumed.

Non-depository — roadmap

CDFI & specialty lenders

Configurable profiles for funder- and state-specific review requirements. One rule throughout: the tool never fabricates a citation.


Data & privacy

Your borrowers’ data, on your terms

This is the first question a security team asks, so here is the direct answer — no marketing around it.

01

Never used to train a model

Reviews run on Anthropic’s commercial API under zero-data-retention terms. Borrower financials are not retained and never become training data.

02

Bring your own API key

Route inference through your own Anthropic account, governed and billed under your contract — your data never touches ours.

03

Bring your own encryption keyEnterprise

Customer-managed keys (BYOK/CMEK) for data at rest. You hold the key; you can revoke it.

04

Encrypted in transit and at rest

TLS on every request, with a signed Data Processing Agreement available for every customer.

05

Multi-tenant isolation

Each institution’s data is strictly partitioned — essential for review firms serving many client banks.

06

SOC 2In progress

Formal SOC 2 attestation is on the roadmap, with the control evidence trail built from day one.


Deployment

Deployment without a project

No install, no infrastructure lift. Sign in and review your first credit the same day.

  • SaaS, today. Runs at the edge for low latency; nothing for your team to host or patch.
  • Bring your own key. Route inference through your own Anthropic account for full data governance.
  • In-your-cloud, on the roadmap. For enterprises that require it, run Claude inside your own AWS (Bedrock) or Google Cloud (Vertex) boundary.
  • Structured intake or upload. Enter a deal, or drop the credit memo and let CRAA spread it.
Analytics

A portfolio you can report on

Every logged review rolls up into a book-level view — the artifacts examiners and your board actually ask for.

Rating migration — illustrative
PassSMSubDoubt
Pass88930
SM1261243
Sub2157013
Doubt001981

Sample data · plus concentration, staleness & override-rate views


For review firms

Built for the firms that run reviews, too

Loan-review and consulting firms aren’t just competition — they’re the sharpest users. CRAA turns a fixed-capacity practice into a scalable one.

More engagements

Same team, higher throughput

The tool does the first-pass grind, so reviewers spend billable hours on judgment — expanding margin without expanding headcount.

New revenue

Continuous monitoring as a service

Offer always-on review between annual engagements — something no manual practice can deliver today.

White-label

Your brand, many clients

Strict multi-tenant isolation per client institution, under your own identity and workflow.


Model risk

A validation record, not a leap of faith

Because a human owns every conclusion, CRAA is a documentation and consistency assistant — not an autonomous decision model. And where scrutiny applies, we bring evidence.

Gold-case evaluation

Tested against real outcomes

A curated set of historical credits across C&I, CRE and stressed segments, run on every prompt or model change to catch regressions before release.

Drift monitoring

Watched over time

Rating-distribution shifts, segment override spikes and hallucination trends are tracked, with defined investigation thresholds.

Change management

Versioned & reversible

Every review stamps the exact prompt, model and regulatory-knowledge version that produced it, with a changelog and rollback path.


See CRAA review one of your credits.

Bring a deal you already know the answer to. In one session you’ll see the rating, the evidence, the narrative — and exactly where your judgment still decides.

Advisory only. CRAA’s recommendations are for informational purposes and require reviewer judgment. The tool does not constitute a formal credit determination and is not a substitute for the independent judgment of a qualified credit professional. Every conclusion is owned by a person, not the software.