HeadlinerAI

Inspiration

I've personally always found news articles hard to interact with. As they have hard to understand language, are too long, and contain excess information. So I decided that a PWA webapp, would be a great solution to inspire the younger generation to stay in touch with trustworthy news sources.

What it does

Our PWA webapp delivers a daily news feed that are related to users interests. During the signup process users select 5 topics (can be very specific or broad user inputted strings), these are then used to fetch news articles relating to these topics. We use OpenAI's GPT4-turbo to summarize the news articles into a small and "gamified" texts. We then display 10 news articles with summaries on the homepage and use web push notifications as to notify users of these updates.

How we built it

We built the application using Next.JS, AWS, Shadcn/UI, tailwindcss, javascript, and NewsAPI, OpenAI API. The main tech stack is Next.JS frontend and backend paired with a AWS Aurora Postgresql DB. Authentication through nextauth, google oauth. PWA and web push notis.

Challenges we ran into

There was a lot of issues with the Postgresql database and integrating PWA. The frontend was decently troublesome as we developed for mobile, using regular web orientated desktop toolings.

Accomplishments that we're proud of

We're proud of the amazing looking frontend, most of our members spent the majority of their time working on the frontend as to make it look the best it could. The backend setup is another amazing accomplishment but the SST stack greatly assisted in making it easier to accomplish.

What we learned

Don't start late on a hackathon, we didn't start working on our project or even thinking of ideas till 11am this morning. So we had pulled an all-nighter to the due date as to get this much progress done.