Get ready for some excitement because Amazon Web Services (AWS) has just unveiled PartyRock, an amazing new playground for AI enthusiasts!

This vibrant platform lets you dive into the world of AI app creation, all powered by the dynamic Amazon Bedrock. PartyRock is not just a tool, it's an adventure into generative AI that's both speedy and incredibly enjoyable. As an AWS Community Builder, I had the pleasure of being able to have early access to this and get on to building some apps. It's the perfect blend of learning and fun for anyone keen on exploring the wonders of AI technology!

To get started, you don't need to have an AWS account or have machine learning expertise; all you need is passion and some wacky ideas!

You can start off by heading into https://partyrock.aws/ to try it out.

You'll see the front page as above, along with sample apps that you can try using directly. But we want to get building our own innovations, don't we? So we can click on Build Your Own App and put down a prompt.

Making the prompt detailed enough so that it knows what to do and what kind of inputs and outputs are expected is important. For example, I thought about what I wanted to do and I wanted to recommend people new restaurants to try out based on where they live or an address that they are at, and have PartyRock recommend them, so I did just that.

Now, you can click on Generate app. And just like that, you have an app ready for you to run!

For me, the result of this was NearbyEats. You can directly put an address or a place and have the best restaurants nearby being recommended to you.

But wait, there's more! You can also edit the text that is used in the app, such as the Widget title and the Placeholder. You just need to click on the configurations button on either of the input or output box, and accordingly modify it.

When you're ready to share your app to the public, you can click on Make Public and Share, and you will receive a link in which you can share you app for anyone to try!

You can also click on Remix which will create a copy of the app, or Snapshot in which you can share your input and output with others to demonstrate how to use your app.

My favorite part, however, is that it can also be conversational. After getting a generated output, you can also have a chat with the AI on the output.

You can ask it to justify why it provided that output, or ask to build on it further.

That being said, what is even more important is to understand that you can also adjust the foundational model being used based on your preferences. You can choose from Claude, Claude Instant, Jurassic-2 Mid, Jurassic-2 Ultra, and Command. As per usual, you will need to experiment with each to find how the prompts differ from one another and which works best for you.

And that's a wrap on this brief run-through of the capabilities of PartyRock and how to utilize it for your own use cases.

Build on and I can't wait to see what you come up with! Do share some quirky ones with me as well; feel free to put them in the comments below!