> Markdown version of [/jobs/ext/4466213-backend-engineer-ai-fintech-startup](https://www.wearedevelopers.com/jobs/ext/4466213-backend-engineer-ai-fintech-startup). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Backend Engineer - AI Fintech Startup

- **Company:** RnB Consultancy
- **Location:** San Francisco, CA, United States (Remote available)
- **Experience:** Experienced
- **Salary:** $200,000.0 - $300,000.0
- **Contract:** Permanent contract
- **Skills:** Artificial Intelligence, Computer Vision, Cursor, Large Language Models, Backend, Agentic-AI, Claude, Virtual Agents
- **Published:** October 12, 2026
- **Apply:** https://www.indeed.com/viewjob?jk=c76a582f253bf65b

## About the Role

* 2-6 years as a professional software engineer
 * Real startup experience (a must)
 * Strong backend and infrastructure skills
 * AI, ML or agent systems you have put into production yourself
 * Claude, Cursor or Codex as part of your daily workflow
 * Clear, concise communication; comfortable with customers and monthly travel
 * Something in your background that stands out
 * Based in San Francisco or ready to relocate, and able to start soon

## Description

A Y Combinator-backed startup in San Francisco is building AI agents for commercial lending. At banks and fintechs, underwriting and servicing a business loan can take months. Their agents cut that to under two weeks: they monitor loan portfolios, request missing documents from borrowers, read whatever comes in (from clean PDFs to blurry scans and handwritten notes) and spread the financials using each lender's own rules, with in-house lending experts checking every result. Commercial lending is a $3T market, and a bank spends around $25k to originate a single loan. Their edge: underwriting models tuned per industry, where general-purpose LLMs still fall short.

With 8 pilots running, they're hiring 3 backend engineers for the founding team to turn those pilots into seven-figure contracts.

What will you do?

 * Own one production AI agent end to end, most likely starting with the borrower portal and document parsing
 * Build the document AI and computer vision that read financial statements
 * Reconcile conflicting numbers across statements in real time
 * Build the backend and infrastructure that take the product from pilot to scale
 * Visit customers on site about once a month and turn what you learn directly into product