Prompt search
Visual search
Output format

The Stylist as a Service

Ever found a t-shirt you liked but couldn't find it anywhere? Do you struggle deciding what to wear when going out? ShopTheLook will be your savior in these situations. Thanks to our AI-powered assistant, you can search and discover clothes from the palm of your hand. Just open WhatsApp and start chatting. Add it to your group chat, and you'll always know what to wear!

What it does

When you send an image of a clothing piece to the chat, the assistant searches for similar items and presents them to you with a big picture, a brief description, a price tag, and a link where you can purchase it. Moreover, if you don't know what you're looking for, you can ask for recommendations given your style and use case.

How we built it

The user interface is a WhatsApp chat that receives message events and forwards them to the backend. There, they're processed by an LLM, and using InditexTech's API, we obtain products that adjust to the given requirements. These results are then sent to the chat, where they are presented to the user in an intuitive manner.

Challenges we ran into

All the process was a constant uphill battle. The unorthodox user interface created some unique constraints when using the provided APIs, but using novel technologies, we were able to overcome them.

Accomplishments that we're proud of

Despite all the challenges stated previously, we managed to deliver a product that exceeded our own expectations, and we are very proud of it. Our unique frontend made us have to search for other solutions when using InditexTech's API, but the solution we reached left us very satisfied.

What we learned

The two less experienced first-year computer science students learned the basics of backend development and API interaction. In addition, the scope of our solution required us to branch out into new fields such as AI, cloud services, and web scraping. We learned to interact with the unofficial WhatsApp API and the OpenAI GPT API. Even though the final product didn't require a database, we also learned MongoDB. Finally, we studied InditexTech's API deeply as it is the main backend tool our product is based on.