Share your submission video!

Adding mine - Tabletop forever. It’s a set of AI agents that write a Dungeons and Dragons campaign from scratch and then play through it.

1 Like

Now I’m sharing our submission(The Blend) here jus check it and kindly leave your feedback

The Blend is a personalized radio platform. With The Blend, you can create your own interactive experience to “cook” the perfect radio mix. Customize your listening journey by selecting and combining various content types, and enjoy a radio experience uniquely tailored to you.

You also have chance to support us by voting here

3 Likes

Have you ever heard me talk about Spreadsheets?
Perhaps my demos about talking with spreadsheets?
If you’ve spent much time with me over the past decade+, you have!

For those that haven’t, one of my goals has been to make tools that make Google Sheets and Docs as easy to use as talking to thin air. Imagine asking the computer on Star Trek a question or to record some data. Whatever you use spreadsheets for - the goal is to make a tool that will help you use them easier and more ubiquitiously, while still putting you in total control.

The Gemini API, and a host of other technologies, have brought this vision one step closer.

Take a look at a quick demo (and amusing story) prepared for the Gemini API Competition.

The system diagram shown there only gives an overview of how everything works together. I discuss this even more with my friend, Mark Tucker, on our Two Voice Devs podcast, where I break down how each part of the system works. We explore

  • exactly where Gemini fits in and the many tasks it does
  • how these tasks are controlled using agents built with LangChain.js and LangGraph.js
  • the dual-asynchronous messaging queues built with Firebase Firestore that are
    key to the conversations
  • where other tools from Google Cloud and Firebase fit into the picture, including authentication and authorization, Google Cloud Storage, TTS and STT APIs, and, of course, the Google Drive API and Google Sheets API.

Take a look at this podcast.

My hope with Vodo Drive is twofold:

  • To bring the flexibility of LLMs with Gemini to the flexibility of Google Sheets
  • To help other developers understand how Gemini, as a tool, can make our existing applications even better and inspire tools that solve the problems, big and small, that impact us every day

If you like what you see - please vote in the People’s Choice awards, and reach out to me so we can chat about what it means to truly have dynamic and living conversations with our data and our world.

3 Likes

Hi! Everyone…
Now disabled person can feel and communicate through this app.

[Lets-Talk]

[NOTE: Excuse for the audio]

1 Like

Wingman AI: Immersive AI in any game and unlimited extensibility with Skills (like Alexa etc.)

https://www.youtube.com/watch?v=qR8FjmQJRGE

Want a world with less waste and more amazing projects? Vote for REM in the Google competition! :rocket: REM is our app that uses AI to connect people with leftover construction materials, so we can prevent them from ending up in the trash! It’s easy to use, just take a photo and you’re done! :recycle: Help us win to build a more sustainable future. Your vote counts! :green_heart:
https://ai.google.dev/competition/projects/rem

Thank you for your support! Nelson y Ricardo Ulloa :chile:
2 Likes

Pretty cool looking. It makes me more curious if you used elevenlabs and suno. So is it voice commanded by elevenlabs or just TTS? What did you use suno for?
I’ve found this repo GitHub - Jofthomas/Everchanging-Quest: The source code of the game I made for the HuggingFace game jam but this is all OpenAI. I think it’s ok that it was developed last year for a 48 hour HughingFace Open Source AI Game Jam, impressive for 48 hours EVERCHANGING QUEST by jofthomas
I assume you converted it from OpenAI to Gemini, I am super curious to compare the two versions. The one running on HughingFace right now is the OpenAI or the Gemini version?

Hi and thanks !

So there is ElevenLabs running when you speak to an NPC ( the credits are at 0 since too much people used it this month :sweat_smile: but it was shared by philipp so I had many testers) and on the top right corner of the village there is a bard that can sing some songs with suno. The API being unofficial : , I deployed it on vercel, but there is some “bugs” making it sometime unavailable, a first call wakes it up and fail and then a second actually works !

You found the first version of this project that I made for the Hugging face game jam ! ( I’m now an HF employee after this ). It was a pygame game using fucntion call from openai to control the game ( more info here ) But pygame having limited option to share your work, I took the oportunity of this contest to start again from scratch in godot (link here) to have WEBGL export available for the space :slight_smile: I indeed changed from OAI to gemini, but also some more ( find backend space here). I also changed the way I represent the generate the maps just for it to be convertible to a markdown representation for the LLM.

I hope it is clearer now :hugs:

1 Like

Wow, very interesting. I’ve never heard of Godot. So it calls Vercel cloud functions? It’s good that you have the PyGame version as well for comparison.

Here is my submission video guys :point_right:https://youtu.be/Z1tWAjA2rpQ?si=gp6dBGH3ex52IeY-

Hello Everyone,
I made a AI Powered drawing game “AI - Playground”.
AI Playground is a groundbreaking game that empowers users to unleash their inner artist through interactive challenges. By harnessing the power of Gemini’s advanced language and image capabilities, we’ve created a platform that fosters creativity, learning, and fun.
Support and Vote here.: AI Playground  |  Gemini API Developer Competition  |  Google AI for Developers

Here is My Video:

3 Likes

This is a really interesting idea!
Do you take advantage of Gemini’s safety filters as part of the community and sharing component?

1 Like

Yup @afirstenberg,
I used gemini’s safety filters and checks to make sure drawings not violates or abuse the community.

Thankyou for letting interest in my project. :+1:

2 Likes

Fantastic! Would love to read more (in a separate thread) about how you’re using it and how you handle the safety responses.

This is something that I feel people aren’t really talking about, and especially in clear use cases like yours, I think it would help the community to learn more.

2 Likes

Sure, I really like to share the detailed explanations about that.
Will write a separate thread soon and let you know about that when it been post.

1 Like

Amazing projects everyone, and the design on these are so cool! Thanks for sharing, it was hard to find projects in the competition, I could only see one random project at a time.

First competition ever for me. I happened to be working on a personal project that was a perfect fit for Gemini when they rolled out the video chat with cache, so I implemented it and joined the competition. It was a lot of fun!

My project is a personal AI soccer (Futbol) coach and analyst. Add your games, generate highlights, and chat with them to get advice and coaching.

Many more features than that, but tried to keep it focused. The video is more of a story of my daughter’s high school state championship season rather than a marketing video. So I doubt the judges got it after seeing some of your really great product videos. But that’s ok, it was about the experience and being a part of something I enjoy that many others I know don’t understand.

3 Likes

Hey all Don’t miss out our demo video too
A space Research assistant
Demo video :

1 Like

Electra AI explores what AI could do for politics:

Issues are generated by reading the news. Then proposals are generated for each issue. You can also chat to Electra about specific issues or policies, or in general about all of them.

I Made an AI powered Social Network, Which is helpful for Blind people too :slight_smile:

Thanks and do share your feedbacks please…

3 Likes