Whether you work in tech or are relatively new to the ecosystem, Generative AI is a buzzword you have most likely come across, especially with tools with ChatGPT. You may not even understand what it is about but now you can!
Check out my workshop on Building 'No-Code' Generative AI Applications with PartyRock✨
AWS PartyRock just launched 🎉 and I think it's a great way for anyone (literally) to learn more about generative AI and AI-powered apps in general. In this post, I share how I used PartyRock to build an app for one of my hobbies: curating skincare routines!
But first,
What is AWS PartyRock?
PartyRock, an Amazon Bedrock Playground
PartyRock is a playground for building generative AI-powered apps, ZERO coding involved.
With PartyRock, you can build your own AI app within minutes and share it with the world! All you have to do is describe what you want to build as your app, and PartyRock does the rest.
This was me 2 days ago!
Being a playground, you get to learn the fundamentals of generative AI in a fun way by building silly (or not-so-silly) apps and experimenting hands-on with prompt engineering.
Note: This is not a guide to learn about Generative AI or prompt engineering, so I'll just include links to resources on them if you want to learn more.
How to Get Started With PartyRock
1. Visit the website and sign up for an account. You can use your Google, Apple or Amazon account.
No, this account is not related to your AWS account, if you have one.
P.S. I like the UI! It gives off a playful feel (which is spot-on for its purpose).
2. Once your account is created, simply get started! You can either:
click on the Build button on the home page,
Scroll to the bottom and describe your app with 'Let's Build', or
Or go to 'My Apps' and click to generate your new app.
You can also remix any of the listed apps if you do not want to start from scratch.
3. Describe your app in 2-3 sentences (or as many details as you can). For mine, I gave some context and what I wanted the app to do.
This gets you started on practising prompt engineering.
Voila! Your app is ready to go!
4. (Optional) Tweak your app
There are several ways to make your app better.
- First, you can try resizing the widgets to get a better layout. This was my final layout after resizing and moving the widgets around.
- You can also add new widgets and edit existing widgets. There are different kinds of widgets as shown below.
- And most importantly, you can edit the AI-powered widgets (the ones that show the results of your app) to improve your prompt engineering skills.
I modified mine to include more details and be more specific with some parts of the responses.
Before:
After
The Result
I also played around with the models and it was interesting to see how they all presented different responses for the part of the prompts that had to do with text formatting.
E.g., Claude Instant sometimes presented the links the way you’d expect (as shown below) but other times it did not. Claude almost always presented links plainly and the Jurassic models ignored all text formatting prompts.
- Claude
Jurassic-2 Mid
And that gives a summary of what you can do with PartyRock.
Check out my app here!
Now, it's your turn!
Try building yourself an app that does anything you can imagine - it could be a daily motivational quotes app or any other random idea you have!
A Question About PartyRock
- No, PartyRock is not an AWS "product". It is simply a playground for you to learn more about generative AI and work on your prompt engineering skills.
What better way to do that than to learn hands-on by building apps?
However, it is powered by the foundational models available on Amazon Bedrock, so you also get to use some of these models and observe how they work before moving over to utilise them in Bedrock.
To learn more about PartyRock and how to use it, check out the following resources:
Thank you for reading and I hope it was a fun one for you.
Like PartyRock says, anyone can build AI apps so, happy building!🎉