Kinnect: Taking the hard out of hardware.
Students, especially young girls, do not often have the money, background knowledge, or resources, or initial confidence to know what kind of hardware they should purchase to learn basic tech skills like circuit development and electronics principles. This is often a barrier to pursuing careers like computer and electrical engineering from a young age.
To overcome this challenge, we built a low-fidelity version of Kinnekt- a gamified lesson in electronics principles that allows players to learn hardware skills without stepping out of the virtual environment.
What it does
Kinnekt teaches students basic concepts in electronics. It then provides players with a challenge mini-project, where they have to draw in the missing components of the circuit to make the project functional virtually. The game uses an Azure based image recognition model that compares the user's answers with the correct solution set and decides whether they have the necessary components to complete the circuit and move on to the next concept.
How we built it
- React Front end using MaterialUI
- NodeJS Backend that accesses Azure Model
- Azure Custom Vision API access model to make comparisons between user solution and correct solution.
Challenges we ran into
Integrating the NodeJS backend with the React Front-end. Training the ML Model. Setting up an aesthetic yet functional UI.
Accomplishments that we're proud of
Building a functional prototype in time. Having a UI of which we are proud.
What we learned
- Integrating so many different components into one project within 24 hours is hard.
- Sleep is essential to coding survival.
What's next for Kinnekt
Setting up more functional cases that with multiple 'project' categories for players to try. Making our ML model more robust.