Built a chatbot trained to explain what AI is to university-educated students.

My Blog

An AI emergency assistant that analyzes text and image descriptions  developed at ConUHacks IX

## Inspiration

The inspiration behind **ResQLens** came from the need to improve emergency response times, especially in high-stress situations where users might struggle to call for help manually. By leveraging AI and hands-free technology, we sought to create a faster and more reliable way to communicate emergencies.

## What it does

**ResQLens** enables users to contact emergency services quickly through Metaglass. Users can send distress signals using voice commands or by taking pictures, which are then processed by AI to generate a clear and concise emergency report. This report is sent via WhatsApp to a backend server, which then relays the message to emergency responders.

## How we built it

*   **PHP** was used to develop the backend server that handles WhatsApp messages.

*   **WhatsApp API** was integrated to facilitate message relays.

*   **ChatGPT** was used for AI-powered summarization of emergency messages.

*   **Metaglass** served as the primary input device for capturing user commands and images.

*   The system was designed to work around Meta's private SDK by leveraging WhatsApp as a communication bridge.

## Challenges we ran into

*   **Lack of an official SDK for Metaglass**, which forced us to find a creative workaround using WhatsApp.

*   **Ensuring message delivery speed** and reliability to make the emergency system effective.

*   **Accurate AI summarization**, ensuring the messages sent to emergency responders were clear and useful.

*   **Integrating AI and WhatsApp** in a seamless and efficient way.

## Accomplishments that we're proud of

*   Successfully built a working prototype despite **Metaglass SDK limitations**.

*   Leveraged **AI to summarize emergencies**, reducing human error and improving response clarity.

*   Found a **practical way to use WhatsApp as a bridge** for emergency communication.

*   Developed a **scalable and adaptable** emergency response solution.

## What we learned

*   **Creative problem-solving** is key when working with hardware that lacks an official SDK.

*   **WhatsApp Business API** can be effectively used for emergency communication.

*   **AI can play a crucial role in emergency response**, making messages clearer and more actionable.

*   **Working as a team** helped us overcome multiple challenges and push forward to completion.

## What's next for ResQLens

*   **Expanding platform compatibility** beyond Metaglass.

*   **Direct integration with emergency services** for faster response times.

*   **Improving AI summarization** for more accurate and useful emergency reports.

*   **Developing an alternative SDK-independent solution** for better Metaglass integration.

*   **Exploring other AI-driven enhancements**, such as predictive emergency assistance.

## Built With

*   chatgpt

*   metaglass

*   openai

*   [php](https://devpost.com/software/built-with/php)

## Try it out

*   [ GitHub Repo](https://github.com/JaedenRotondo/ConUHacksIX)

