How to use Unknown to streamline mortgage processing
SimplAI reports agents handling mortgage documents, income analysis, verification, underwriting preparation, and compliance checks, with production processing falling from around 18 days to three to five.
news.ycombinator.comPrompt to try
Editorial adaptation, not the source author’s exact prompt. Replace bracketed inputs and review tool permissions first.
Process [sample mortgage file] in [approved environment]: classify documents, extract fields, analyze [income records], run [authorized verification integrations], and prepare underwriting conditions. Check against [applicable rules] and present the file, uncertainties and supporting documents to [human underwriter] for review.
Reach people exploring AI agents
See placements →What happened
The team reports processing falling from around 18 days to three to five days, extraction accuracy above 97%, review time under 45 minutes, and approximately 40–50% lower cost per loan.
Reported by SimplAI_ai on Hacker NewsHow the agent did it
- Classified loan documents and extracted their data.
- Parsed tax returns, payslips and variable income records.
- Performed real-time employment and financial verification checks.
Where a person stepped in
The source retains human underwriters and reports review time, but does not describe detailed human review procedures.
A person approved each step before the agent continued.
The SimplAI team described building agents for the operational work surrounding mortgage underwriting. They identified lengthy processing timelines, substantial document handling, repetitive data entry and verification, and underwriters spending hours on work outside the decision itself. The system was designed for banking and financial services, beginning with mortgage operations.
Its Document AI component classifies loan documents and extracts their data in minutes. Income-analysis models parse tax returns, payslips, and variable income. Verification integrations perform employment and financial checks in real time. An AI-assisted underwriting stage pre-validates files and generates conditions, while a compliance engine continuously checks against regulatory rules.
The team reported production processing falling from around 18 days to three to five days. They also reported extraction accuracy above 97%, underwriting review time falling from three to four hours to under 45 minutes, and cost per loan falling approximately 40–50%. The post describes retaining underwriters and removing operational bottlenecks around them. It does not provide the measurement methodology or details of review procedures. The team said edge cases and explainability remained areas of exploration. Who it's for: mortgage operations teams evaluating document and verification automation around human underwriting.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · SimplAI_ai
Frequently asked questions
What work did the mortgage agents handle?
The described components classified and extracted loan documents, parsed income records, performed employment and financial checks, pre-validated underwriting files, generated conditions, and checked regulatory rules.
Did the team say it replaced underwriters?
No. The team said it was removing operational bottlenecks around underwriters, and reported that underwriting review time fell from three to four hours to under 45 minutes.
More details
Remaining steps
- Pre-validated underwriting files and generated conditions.
- Continuously checked files against regulatory rules.
About this account
The agents handled operational document, income, verification and file-preparation work around human underwriting.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=47425482. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: SimplAI_ai on Hacker News
Source published 18 Mar 2026
Source checked 14 Oct 2026
Tools and topics
SimplAI · Document AI · Income analysis models · Verification integrations · Compliance engine
Mortgage operations · Document extraction · Verification
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.