Inspiration
At Imersian, we've had the opportunity to engage with hundreds of homeowners undergoing the journey of furnishing or renovating their homes. We identified a common challenge among them: the overwhelming variety of products, design styles, and brands available. Despite this plethora of choices, homeowners often find it difficult to create spaces that truly reflect their individuality. This insight led to the creation of Immy, designed to simplify the design process and provide a personal design companion for everyone.
What it does
Immy is an AI assistant that delivers design suggestions tailored to users' personalities and recommends actual products that align with these design ideas.
How we built it
Understanding individual personalities can be complex. However, we discovered Vana, an API that develops personality profiles based on people's behaviors and interests. We integrated Vana's insights into Immy, enabling it to offer tailored design recommendations.
While integrating with the Vana API, we compiled thousands of products and categorized them with our machine-learning algorithm.
After these initial phases, Immy has become fully equipped and ready to assist users in their design endeavors.
Challenges we ran into
We faced challenges in classifying products by deciphering their metadata. Additionally, vectorizing data and integrating it with design recommendations proved complex. Understanding users' contexts as they interacted with Immy was another challenge.
Accomplishments that we're proud of
Immy now successfully provides design recommendations influenced by users’ personalities and can analyze an image of a space to offer targeted design enhancements.
What we learned
The development of Immy has been an extensive learning experience for our team. We've gained valuable insights into creating an AI assistant and training a large language model to address specific real-world use cases.
What's next for Immy - your design companion
Soon, Immy will generate images incorporating real-world products, further assisting users in effectively visualizing design recommendations.
Built With
- ai
- amazon-web-services
- computer-vision
- node.js
- python

Log in or sign up for Devpost to join the conversation.