AI News Aggregator

Inspiration

I’ve always been passionate about following the latest developments in AI: from new models and research papers to policy implications and economic impact. But most news aggregators I’ve tried feel overwhelming, cluttered, or filled with surface-level content. I wanted a tool that gave me high-quality AI news from expert sources, in a simple, scannable format. And I wanted to see if I could build it myself, just for fun.

What I Learned

This was my first hackathon, and I’m not a coder by background. I’m a product guy. But this experience taught me how accessible it’s becoming for non-engineers to build meaningful tools.

How I Built It

I used the Perplexity Sonar Pro API to fetch the most recent AI news from trusted sources like research labs, think tanks, and technical publications. I wrote a custom prompt to filter out shallow or paywalled content, and return results in structured JSON format. I then built a minimal web app that calls the API on demand and displays the top articles in a clean, skimmable layout. The backend runs on Node.js, and the frontend is a basic HTML/CSS app optimized for simplicity.

Challenges I Faced

The project isn’t technically complex. I admit that. I also had to troubleshoot caching issues during development and ensure the API integration behaved consistently with each refresh. Debugging JSON responses and making sense of dynamic web content was tricky but rewarding.

Final Thoughts

I built this project in just a few hours, and it’s already become a tool I use daily. I’m incredibly thankful to Perplexity for offering this kind of API. It lets people like me experiment, learn, and build useful tools quickly. I believe this is the future of AI: individuals building their own apps on top of intelligent APIs.

Built With

• bolt.new
• css
• express.js
• html
• javascript
• node.js
• perplexity
• sonar