Newsify - AIPhotoJournalist

Main menu, this is where the user can input an image, and select a style of article
Output of an image. Shows response and dynamic css, as well as a clear view of the image
Branch history of all the versions of the code during the hackathon
Website is also suitable for mobile users thanks to the dynamic css
Website main menu also works with mobile users, making sure everything is laid out correctly

Inspiration

Inspired by the UniCS challenge for PhotoJournalist, and also by the AI Agent Reply challenge. Therefore creating an AI PhotoJournalist

What it does

Takes either user input, or camera input. User selects style of article, then waits for a response. Dynamic CSS changes web layout, and generated text about the image is given. The user can then go back and submit another image

How we built it

React js for the front end, which handled HTML, CSS and JSX to make the UI look amazing Python Flask for back end, which allowed us to use a multitude of Python modules, include openai

Challenges we ran into

At first, we used Svelte for front end, but after talking to a mentor a couple hours in, we were steered towards using Native React JS (since it was a lot better for complex, and multiplatform projects)

We also tried using the Google Vision API, but it only gave tags rather than a good response.

Then, we had to learn React, since it was new to all of us We also had to learn flask and how to implement the openapi key And then we had to learn how to link the two together Having to style all the css files to fit both desktop and mobile

When implementing dynamic CSS and images, the file directories were very fussy Since we used React Vite instead of Expo, it was a lot more challenging to try and implement a web app

Finally, testing it on each of our systems to make sure it didn't only work on one device, which required the use of a gitlab repository, now a github repository

Accomplishments that we're proud of

Able to build a working project with frameworks and libraries we've never used before Able to use AI in our project Able to make our project look sleek and presentable We attempted and tried to achieve every extension challenge

What we learned

How to work well as a team, we split the team up in a way where we were able to create large chunks of the front end and back end simultaneously That anything is possible if you put your mind towards it. In the first few hours, I said "How can I learn React in one Hackathon", and now we have a fully working project.

What's next for GreatUniHack - UniCS Challenge - Team YFive.

First year team project Implement frameworks into own projects and school projects Very proud in front end and linking it with backend May join future hackathons