🧠 MedGuide AI — Autonomous Drug Interaction Intelligence

Built with AWS Bedrock · OpenFDA · Serverless Architecture

MedGuide AI is an autonomous, AI-powered drug interaction platform built to prevent harmful medication combinations through advanced reasoning, verified data, and secure, QR-coded clinical reports — all in under 10 seconds.

---

🚀 Overview

Every year, over millions of hospital admissions occur due to medication errors. MedGuide AI was built to change that — by empowering doctors and patients with real-time, data-driven drug safety intelligence.

This project was developed as a solo end-to-end build for the AWS AI Agent Global Hackathon, integrating AWS Bedrock AgentCore, Nova reasoning models, and OpenFDA APIs within a fully serverless architecture.

---

🩺 Key Features

• AI-Driven Drug Interaction Analysis — Uses AWS Bedrock’s Nova model for reasoning over complex drug combinations.
• OpenFDA Verified Data — Pulls real-time clinical and pharmaceutical data from FDA sources.
• Comprehensive Interaction Reports — Generates structured, color-coded reports with clinical significance and recommendations.
• QR-Based Verification System — Each report includes a unique QR for secure mobile access and verification.
• Serverless AWS Stack — Lambda + API Gateway + S3 + IAM + KMS for seamless, secure scalability.
• Modern UI/UX — Fully responsive, minimalistic web interface built for clinicians and patients alike.

---

🧩 System Architecture

Patient → CloudFront → API Gateway → AWS Lambda → AWS Bedrock (Nova)
                                          ↓
                                  OpenFDA APIs (Drug Data)
                                          ↓
                              AI Reasoning & Report Generator
                                          ↓
                              Encrypted S3 (Report Storage)
                                          ↓
                               QR-Verified Clinical Report

• Front-End: React + Tailwind CSS
• Backend: Node.js (Express)
• Database: Serverless storage via AWS S3
• AI Layer: AWS Bedrock (Nova) + AgentCore
• Data Source: OpenFDA API

---

⚙️ How to Run Locally

1. Clone the Repository

   git clone https://github.com/yourusername/medguide-ai.git
   cd medguide-ai

2. Install Dependencies

   npm install

3. Add Environment Variables Create a .env file in the root directory:

   BEDROCK_API_KEY=your_aws_bedrock_key
   OPENFDA_API_KEY=your_openfda_key
   S3_BUCKET=your_s3_bucket_name

4. Start the Development Server

   npm run dev

5. Access the App Open your browser and visit: http://localhost:3000

---

🧠 Built With

• AWS Bedrock (Nova)
• AWS Lambda
• AWS API Gateway
• AWS S3
• AWS IAM & KMS
• OpenFDA APIs
• Python

---

🧍‍♂️ Project Lead

👨‍💻 Developer: Azkhan

---

🩶 Vision

“The last revolution in patient safety was electronic health records. The next will be autonomous AI agents safeguarding every prescription. That’s what MedGuide AI was built to do.”