Wardrobe Whiz

• 

  Wardrobe Whiz

Inspiration

Laziness. I'm sorry, but that is the best way I can describe it. It takes a lot of time for myself, as a perfectionist, to pick out the right clothing to wear. Often there is a bit of anxiety associated with the process as well. Fashion comes naturally to some people, and they can make anything work. So, this is the app for people like me who are challenged by the idea of having to dress themselves for any occasion so often.

What it does

It helps to take the anxiety out of picking the right outfit for any occasion by offering suggestions and displaying an image of the suggested style so that the user can determine if it is right for them and give them the confidence to face the day.

How we built it

Using Amazon Bedrock with the people in mind

Challenges we ran into

There were some challenges with not overwhelming the user with prompts and questions before the outfits were generated. I want the apps I create to be as easy and as accessible to as many people as possible. So the challenge was taking in as little input as possible while still being able to achieve acceptable output and not overload the Gen AI populated field prompts. A future feature may include a dedicated model to items that the user already owns. For now though, the chatbot is a substitute for descriptions of what can be added to complete the look based on what the user indicates they own.

Accomplishments that we're proud of

I am proud of being able to take this concept from an idea to a functional app. I feel this app is simple to use even if the planning was challenging. The prompts I used to focus the app on tasks increased my understanding of how to communicate with AI to get more desirable results.

What we learned

I learned more about the user journey and prompt engineering.

What's next for Wardrobe Whiz

I hope to expand on this idea in mobile app development. I will revisit this to see if I can compete in the world of everyday fashion possibly in a more sustainable materials and practices oriented marketplace.

Built With

• amazon-web-services