Lead Software Engineer
Completed

ANS: AI Platform for Large-Scale Reimbursement Analysis

I led the development of a configuration platform that enabled an AI engine to analyze over 1,000 healthcare reimbursement claims per day with enhanced accuracy and transparency.

Client
ANS (Brazil's National Health Agency), via Paipe software house
Role
Lead Software Engineer
Period
2024
ANS Logo

The Challenge

Understanding the business context and identifying the core problems that needed solving.

The Client

ANS (Brazil's National Health Agency), via Paipe software house

A leading organization in their sector, facing significant operational challenges that required innovative technological solutions to maintain competitive advantage and operational efficiency.

The Problem

Paipe had developed a powerful AI for claims analysis, but it operated as a 'black box', making it impossible for analysts to understand its decisions or make real-time adjustments.

This created significant bottlenecks, increased operational costs, and limited scalability potential.

My Objectives

Given the challenges identified, I was tasked with delivering specific outcomes:

1

Enable fine-tuning of the AI's decision-making rules

2

Provide a real-time performance and accuracy dashboard

3

Ensure full auditability and transparency for every AI-powered decision

2
Quantitative Results
3
Qualitative Outcomes

The Solution

My strategic approach to solving the challenges through innovative architecture and technology choices.

Technology Stack & Reasoning

React

Chosen to build a complex and highly interactive UI, including a real-time dashboard and a visual editor for complex decision trees.

Node.js

Selected to create a lightweight and fast API layer, acting as an efficient bridge between the user-facing configuration app and the core AI engine.

Key Features Implemented

1

Visual Decision Tree Editor

An intuitive graphical interface allowing analysts to view and modify the AI's decision-making logic without writing any code.

2

Real-time Accuracy Dashboard

Displayed live metrics on the AI's performance, success rates, and edge cases, providing immediate feedback on configuration changes.

3

Detailed AI Analysis Log

A comprehensive and searchable log of every decision made by the AI, ensuring a complete audit trail for compliance purposes.

Results & Impact

The measurable outcomes and lasting impact achieved through strategic implementation and execution.

Quantitative Results

1

Enabled the analysis of over 1,000 claims per day

2

Allowed real-time fine-tuning of the AI model, ensuring continuous accuracy against regulatory updates

Qualitative Impact

1

Increased analyst trust and adoption of the AI tool

2

Provided full transparency into the 'why' behind each AI decision

3

Greatly improved the ability to audit and comply with regulations

Interested in Similar Results?

I'd love to discuss how I can help bring similar success to your next project. Let's explore how we can work together to achieve your goals.

View All Cases