FinTech AI SaaS: FastAPI + SQLModel + React/Vue

## Job Description## Overview
We are seeking an experienced, high-velocity Full-Stack Developer to build and launch an AI-driven credit intelligence and automated dispute generation platform.

The architecture, database schemas, API data contracts, and deployment workflows are already fully designed and documented. We need a sharp developer who can take our structured technical blueprints and build a stable, production-ready MVP within a 30-to-45-day window.

This is a clean-slate project with zero technical debt. You will be working with a lean, modern async Python stack on the backend and your choice of a clean SPA framework on the frontend.
------------------------------
## The Technical Stack

* Backend Framework: FastAPI (Python 3.11+, fully async, Pydantic v2)
* Database & ORM: PostgreSQL + SQLModel (SQLAlchemy 2.0 async core) + Alembic for migrations
* Background Workers: ARQ (Redis-based asyncio-native task queue)
* AI Integration: Anthropic Claude 3.5 Sonnet API (Native SDK with structured Pydantic tool calling)
* Document Generation: WeasyPrint + Jinja2 (HTML/CSS-to-PDF compilation)
* Infrastructure & Deployment: Docker + Managed PaaS (Railway or Render) + Cloudflare proxy/WAF
* Third-Party APIs: AWS Textract (tabular extraction), Stripe (Subscriptions/Webhooks), Postmark (Transactional email)
* Frontend: React (TypeScript) or Vue.js (Developer choice, must match provided JSON data contracts)

------------------------------
## Key Responsibilities & Weekly Deliverables

* Week 1 (Ingestion): Set up local Docker containers; build the SQLModel database tables; implement secure PDF upload to S3; hook up the ARQ worker to pass files to AWS Textract.
* Week 2 (Core Logic): Build the rule-based scanning logic to flag credit report errors; wire up the Anthropic SDK to generate structured legal narratives; configure WeasyPrint to output print-ready dispute letter PDFs.
* Week 3 (Integrations): Integrate Stripe subscription tiers and webhook endpoints; connect Postmark transactional email notifications; deploy the dual-process container stack to a managed cloud PaaS.
* Week 4 (Hardening & Launch): Implement security headers, rate limiting, and log scrubbing to protect PII; write basic async integration tests with Pytest + HTTPX; ensure the frontend perfectly hooks into the async polling logic.

------------------------------
## Required Qualifications

* Expert-level Python proficiency, with deep experience writing pure async code in production.
* Proven experience building production-ready REST APIs using FastAPI and Pydantic v2.
* Strong understanding of relational databases (PostgreSQL), query optimization, and handling JSONB data.
* Familiarity with background processing systems (ARQ, Celery, or RQ) and Redis.
* Experience formatting documents using HTML/CSS for print media (experience with WeasyPrint is a massive plus).
* Comfortable working with multi-stage Dockerfiles and pushing code to PaaS environments like Railway or Render.
* Clean, highly structured git workflow and clean code architecture layout principles.

------------------------------
## Project Details & Budget

* Project Length: 4–6 weeks (30–45 Days maximum target window)
* Project Type: Fixed-price with clear milestone payouts (or hourly for the right verified expert)
* Availability: Looking for someone who can dedicate substantial focus to this project immediately.

------------------------------
## To Apply
Please include a brief summary of a production-level async Python application you have built from scratch. Bonus points if you link to a public GitHub repository showing your clean backend structure, Pydantic integrations, or Docker setups.

Please use the phrase "READY TO PARSE" at the top of your application so we know you read the complete project requirements