BalansBuddy

The Problem

People in debt are already at their most vulnerable — and the process asked them to spend 50+ hours on paperwork per application before they could even get help. The existing digital "solutions" were chatbots: they answered questions about the process but did none of the actual work, leaving all 50 hours firmly on the applicant's shoulders. The result was a 55% rejection rate — more than half of all applications failed, usually not because people didn't qualify, but because a single form was incomplete or one document was wrong. For someone drowning in debt, a rejection on a technicality can cost months.

• 50+ hours of administrative work per application
• 55% rejection rate on debt relief applications
• Chatbots that answered questions but didn't do the work
• No multi-language support for diverse populations

The Solution

We built BalansBuddy: a multi-portal platform serving the three parties in every debt case — the client, the counselor, and the municipality — with Ember AI as an autonomous agent at its core. The distinction matters. Ember is not a chatbot bolted onto a form; it is an agent that does the work a caseworker would otherwise do by hand: handling creditor correspondence, reading and analyzing bank statements, and detecting the legal escalations that turn a manageable case into an emergency.

Ember AI doesn't just respond to questions. It proactively handles creditor emails (<5 min response, 24/7), analyzes bank statements to detect patterns and income flows, recognizes 50+ known creditors automatically so their correspondence is routed and answered correctly, and flags legal escalations — like an impending seizure — before they become problems. And it does all of this in 6 languages, because the people who need debt help in the Netherlands don't all speak Dutch.

Built for security and scale from the ground up: React, TypeScript and Node.js on the application layer, PostgreSQL for data, OpenAI GPT-4 driving Ember, and Redis/BullMQ handling the background jobs that let the agent work around the clock. Because the platform handles some of the most sensitive data there is — people's entire financial lives, under Dutch and EU privacy law — every sensitive field is protected with AES-256-GCM end-to-end encryption, and every action Ember takes is logged and reviewable rather than opaque. Privacy and auditability were requirements from the first line of code, not features added at the end.

The Results

The impact was immediate and measurable — and it landed on both sides of the desk, for the people applying and the counselors helping them:

• 98% time reduction: Application time dropped from 50 hours to 1 hour
• Rejection rate 55% to 5%: Complete, accurate applications through AI assistance
• Creditor response <5 min: 24/7 automated correspondence handling
• Counselor capacity doubled: Focus shifted from paperwork to people

Why This Matters

The lesson wasn't about technology — it was about the gap between answering questions and doing the work. BalansBuddy proves an AI agent can autonomously handle complex, high-stakes workflows: creditor correspondence, document analysis, legal escalation detection — with every action logged and reviewable, in six languages, behind military-grade encryption. That capability is directly relevant for clients who need:

• AI agents that autonomously handle complex workflows
• Multi-portal platforms with role-based access
• Sensitive data handled with enterprise-grade security
• Measurable impact: 98% time reduction, 10x fewer rejections

If you need an AI solution that doesn't just answer questions but actually does the work — we've built it.