惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

D
Docker
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Last Week in AI
Last Week in AI
博客园_首页
Microsoft Security Blog
Microsoft Security Blog
Blog — PlanetScale
Blog — PlanetScale
M
MIT News - Artificial intelligence
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
A
About on SuperTechFans
aimingoo的专栏
aimingoo的专栏
V
Visual Studio Blog
Jina AI
Jina AI
N
Netflix TechBlog - Medium
量子位
博客园 - 三生石上(FineUI控件)
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
I
InfoQ
J
Java Code Geeks
T
Tailwind CSS Blog
博客园 - 司徒正美
Stack Overflow Blog
Stack Overflow Blog
阮一峰的网络日志
阮一峰的网络日志
Engineering at Meta
Engineering at Meta
腾讯CDC

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
I built Reclaim: an AI tool that finds medical billing er...
Isaiah Kim · 2026-06-14 · via DEV Community
Cover image for I built Reclaim: an AI tool that finds medical billing errors and writes your appeal letters

Isaiah Kim

Americans owe $220 billion in medical debt, and 80% of medical bills contain errors — duplicate charges, upcoded procedures, unbundling, wrongful denials, balance-billing violations, out-of-network surprises. I built Reclaim to catch those errors and handle the appeals process end to end.

What it does

You upload a photo or PDF of your medical bill or Explanation of Benefits. Gemini 2.5 Flash extracts every line item, CPT code, and charge. Reclaim then checks for:

  • Duplicate charges
  • Upcoded procedures
  • Unbundling violations
  • Wrongful claim denials
  • Balance-billing violations
  • Out-of-network surprises

If no errors are found, you pay nothing. If errors are found, a one-time $29 per-case unlock generates the appeal letter and sends it directly to your insurer's fax line via the Telnyx Fax API — with a delivery receipt and a 30-day response calendar.

Why I built it

Most people don't know they can appeal a medical bill, and fewer know how. The process requires specific legal language, relevant policy citations, and tight deadlines. I wanted to automate the parts that make it inaccessible: document parsing, error detection, letter drafting, fax sending, and deadline tracking.

How it works

The intake flow accepts photos or PDFs. Gemini 2.5 Flash handles extraction. Supabase Postgres stores cases under row-level security per user, with medbill_*-prefixed tables. Auth is Google OAuth via Supabase Auth.

After errors are confirmed, Stripe Checkout handles the $29 unlock. The app generates the appeal letter with pdf-lib — supporting internal first-level appeals, external review requests, and FDCPA cease-and-desist letters depending on the situation — then faxes it via Telnyx with a delivery receipt.

A Vercel cron job runs daily and sends reminder emails via Resend at 7 days, 3 days, 1 day, and overdue per appeal.

The dashboard uses what I'm calling a brutalist editorial design: black on cream, no rounded cards, typography-forward.

Stack

Layer Technology
Framework Next.js 16 (App Router) + React 19
Language TypeScript
AI Gemini 2.5 Flash via @google/generative-ai
Database Supabase Postgres
Auth Supabase Auth (Google OAuth)
Payments Stripe Checkout
Fax Telnyx Fax API
Email Resend
Cron Vercel cron
PDF pdf-lib + pdf-parse
Styling Tailwind CSS v4

Pricing

Upload and analysis are always free. The $29 unlock only fires after errors are confirmed. 100% of any recovery is yours.

Repo: https://github.com/kyisaiah47/reclaim (MIT)