PrivateLeftStoat

Inspiration

Our inspiration was our mutual love for Monty Python sketches.

What it Does

The app takes user's idea for a Monty Python sketch generates a corresponding image and a script for that episode.

Technologies and How We Built It

The front-end was built using HTML, CSS and JavaScript. In the back-end, we used Python for stable-diffusion integration, Google Cloud and Vertex Generation, and OpenAI's ChatGPT.

Challenges we Ran Into

For the front-end side, none of us knew how to create CSS animations so visualising our idea using the code was a bit finicky. Our front-end developer had never focused on vanilla JavaScript before with a main focus in TypeScript and React, so it was a bit of a challenge for them to go back to basics!

On the back-end side, typical issues with Large Language Models were seen. API issues ensued as would be expected and at one point we had to intercept our API traffic to troubleshoot.

Accomplishments that We're Proud Of

For two of our team members it was their first hackathons. One was particularly proud of managing the majority of the front-end by themselves, deepening their knowledge around front-end libraries and allowing them the chance to expand their skill set. The other found it particularly fun continuing to wrangle ChatGPT outputs. Our other two developers enjoyed the chance to continue their personal development across multiple technologies with one interfacing with the stable-diffusion library and the other learning how to interface with Google Vertex.

All in all, our team had a wonderful time deepening our technical knowledge and learning new technologies.

What we learned

We learnt how to create animations using CSS and JavaScript and how to set up local generative AI's.

What's next for PrivateLeftStoat

Hopefully, many more projects like this! We worked great as a team, and we hope we are going to once again be able to come together again in the future for some coding fun.