DEV Community

Congrats to the Cloudflare AI Challenge Winners!

dev.to staff on April 16, 2024

Today is the day! We are so excited to announce the winners of our first-ever sponsored challenge: The Cloudflare AI Challenge. We were completely...

Read full post

Merbin J Anselm • Apr 16

Thank you dev.to and Cloudflare team! I thoroughly enjoyed the challenge. My wishes to @kaarthik108 and @julianandreszb! Congratulations!

Ben Halpern • Apr 16

Congrats!

Merbin J Anselm • Apr 16

Thank you! :)

Nabil Alamin • Apr 16

Congratulations 🎉, really loved your submission

Merbin J Anselm • Apr 17

Thank you! I loved your submission too! :)

julianandreszb • Apr 16

Hi guys, I am very happy to have participated in the challenge! I learned a lot and saw fascinating projects. Thanks to @thepracticaldev and @cloudflaredev for this challenge! Best wishes to @anselm94 and @kaarthik108! And a special thanks to the entire developer community.

Ben Halpern • Apr 16 • Edited

Congrats everyone, there were sooo many excellent, qualifying projects. And I learned a lot just reading through them.

Peter Saulitis • Apr 16

The Cloudflare team was blown away by how well you leveraged Workers AI and our developer platform to build INCREDIBLE AI-powered experiences.

We'll be sharing more of our favorites on Twitter, so follow us to stay tuned.

Huge shout to the Dev.to team, who did an amazing job hosting and running this challenge!

Let us know if you want to do more Cloudflare challenges! Multi-player app, perhaps??

Ranjan Dailata • Apr 16 • Edited

@saulitis Of course, we can have multiple challenges. However, please provide a full trial account with all the features including the Vector DB creation etc. using the trial account. This is one of the crucial thing which was lacking and because of which many of us weren't able to integrate or perform the RAG for this challenge with the trial account. Apologize if my understanding is not right.

It would be great if there is a way to post all our feedbacks on the AI model consumption. At-least from what I have noticed, the worker AI was failing in some scenarios as the LLM wasn't able to handle it. There were Summarization models, it never worked as expected with the input max token as 1024 something context window :(

Peter Saulitis • Apr 17

@ranjancse This is fantastic feedback, thank you so much. I've sent our product team a note on the AI model consumption / errors and will get back to you ASAP.

Per the trial account - we're actively evaluating solutions to this. To best support these types of amazing challenges, we need an easier way for participants to access more of the developer platform features.

Thank you, again, for taking time to write and send this feedback.