Flare.ai

Inspiration

Who hasn't had that moment, standing in front of a bursting wardrobe with absolutely no idea what to wear? And it's not like we don't have the clothes - we do! It's just figuring out what goes with what that takes so much time and energy. We realized most of us get stuck in a fashion rut, not because we want to, but because we just can't find the time or brainpower to mix things up. That's how Flare came into being. We wanted to make it easy for people to discover new ways to wear what they already have, and in doing so, bring a dash of fun and excitement to their everyday fashion game.

What We Learned

Building Flare has been one heck of a ride! We really dug deep into JavaScript, learning the ins and outs of React.js for front-end stuff, Node.js and Express.js for the back-end. The cherry on top was getting our hands dirty with TensorFlow.js to bring in the whole image recognition magic. We didn't just learn about the tech bits, but also how important it is to understand what our users really need.

How We Built It

Flare is all about JavaScript, through and through. We crafted the front-end with React.js because it's just so darn good for building user interfaces. For everything going on behind the scenes, we brought in Node.js and Express.js. To make Flare really see and understand your clothes, we powered it up with TensorFlow.js. We store all the fashion and user data in MongoDB, and AWS S3 is where all your clothing images go. We've got everything up and running on an AWS EC2 instance.

Challenges We Faced

Getting Flare to actually recognize and analyze clothing was one tough cookie! TensorFlow.js and machine learning were pretty much new territories for us, so there was a lot of learning on the job. Making sure Flare was easy to use and actually helpful was another biggie. We had to balance all the techie stuff with creating a cool user experience, which meant loads of tweaking and tinkering based on user feedback.

The Journey Ahead

We're stoked about how far we've come with Flare, but we know it's just the start. We're brimming with ideas to make our AI smarter, grow our fashion database, and add more nifty features for our users. Watch this space!

Built With

• database
• javascript
• python