Imagine you’re planning a grand adventure to Las Vegas and the iconic national parks of Arizona and Utah, envisioning a journey filled with the vibrant life of the city and the serene beauty of nature. Yet, the complexity of planning such a diverse trip can be daunting. What if there was a way to streamline this process? What if you could create an application that simplifies planning any trip, where all you need to provide are some basic pieces of information like the destination, the number of days you wish to spend, and your budget?

This is not just a hypothetical scenario anymore. With AWS PartyRock, such an application is not only possible but can be created in a matter of minutes without writing a single line of code. AWS PartyRock, a platform designed to democratize the creation of generative AI applications using Large Language Models (LLMs), enables anyone, regardless of their technical background, to build powerful tools that can tackle complex tasks like trip planning.

The ease of use of AWS PartyRock is one of its most compelling features, making it an accessible tool for individuals of all skill levels. With its simple and intuitive interface, users are guided through the process of creating applications with minimal effort. One of the standout aspects of PartyRock is its built-in models, which significantly streamline the development process. Rather than getting bogged down in the complexities of coding, users can focus on describing their requirements for the app. Whether it’s a trip planning application or any other tool, you simply outline what you need — for instance, a trip planner that considers destinations, duration, and budget — and AWS PartyRock takes care of the rest. It automatically generates a foundational version of the application based on your specifications, ready for you to refine and customize. This revolutionary approach not only saves time but also opens up the world of app development to those who may have been intimidated by the technical barriers of the past. With AWS PartyRock, creating custom applications is no longer a daunting task but an enjoyable and fulfilling experience.

You start with describing what you want your app to do.

PartyRock automatically create a fully functional app for you. In the app it generated for me, it has three input widgets taking input from user. It has a recommendation widget to show the recommended places and an itinerary widget to show day-by-day itinerary. It also has a chatbot you can talk to it freely.

Let’s give it a test!

The app is fully functional, and the results are impressive! AWS PartyRock adeptly generated prompts tailored to the specific needs of my app, seamlessly integrating the input parameters within these prompts. Additionally, the widget is customizable, allowing modifications to the title, model, prompt, and even the advanced settings to better align with my preferences.

You can also create a widget by clicking “Create Widget”.

“A widget is a UI element that you can combine with other widgets to create an app. Widgets display content, take in input, connect to other widgets, and create output. Widgets that take in input allow users to interact with the app. Widgets that create output use prompts and references to other widgets to generate something like an image or text. The widgets that create output in PartyRock are AI-powered. AI-powered widgets are based on foundation models to utilize AI content generation.” — PartyRock Guide

“There are 3 different types of AI-powered widgets: image generation, chatbot, and text generation. You can edit AI-powered widgets to connect them to other widgets and make their output change.

There are also 2 other widgets: user input and static text. The user input widget allow users to change output when you connect it to AI-powered widgets. The static text widget provides a place for text descriptions.” — PartyRock Guide.

Prompt engineering is key. Prompt engineering has emerged as a crucial skill in the era of generative AI, serving as the bridge between human intent and machine understanding. This discipline involves crafting questions or prompts in a way that guides AI models, like those utilized in AWS PartyRock, to produce desired outcomes with higher precision and relevance. The significance of prompt engineering cannot be overstated, as the quality of the input directly influences the quality of the output. Effective prompts can unlock the full potential of AI, enabling it to generate solutions, ideas, or content that closely aligns with the user’s needs.

I chose to enhance the user experience in the trip planner app by incorporating additional input fields such as “number of people,” “time of trip,” and “departure city.” Instead of dividing the plan across two separate widgets, my goal was for the LLM to produce a singular, detailed plan that encompasses multiple aspects, including:

Furthermore, I integrated a chatbot feature to provide instant answers to any user inquiries. To ensure the chatbot maintains the context of the conversation, I included all relevant background information and the trip plan generated by the app in its prompt. Lastly, I introduced an image generation widget, designed to create cinematic images based on the “place” and “time of trip” specified, adding a visually engaging element to the planning experience.

Here’s the trip plan crafted by the App, presented in markdown for a beautifully rendered display within the app.

Here is a comprehensive trip plan for your 7 day trip to Las Vegas and national parks in Utah, NV, and AZ with a budget of $5000 for 4 people in June:

Trip Planner aims to make trip planning accessible and stress-free for everyone. By personalizing itineraries based on user preferences, budget, and duration, this application can significantly enhance the travel experience, encouraging more people to explore new destinations. Its real-world application lies in promoting tourism and cultural exchange, potentially boosting local economies. Encouraging adoption through travel blogs, partnerships with tourism boards, and social media campaigns can highlight its value in planning memorable, hassle-free trips.

Had AWS PartyRock not been available for developing the Trip Planner app, turning to Amazon Bedrock as an alternative would involve a more hands-on approach to leveraging AWS’s suite of services and machine learning capabilities. Amazon Bedrock provides a solid foundation for building apps with LLM, offering flexibility and control over the development process.

Explore everything firsthand through the following links: