Inspiration

We were inspired by the simplicity and modernity of Valorant's UI and design elements, which we sought to emulate in our app.

What it does

VCT Valiance is an AI chatbot trained on Valorant's pro scene. It provides accurate and insightful information about the Valorant's pro scene and its players. VCT Valiance also has the ability to generate and store custom teams, created as per a user's specifications.

How we built it

VCT Valiance was written using the following technologies:

Challenges we ran into

The main challenge we faced was the limited capabilities and documentation on Kotlin Compose's WASM compilation target. Currently, KCM for WASM is in its Alpha stage, meaning it is functional, but has very limited functionality and documentation. This made debugging and initial framework development very difficult, as we had to learn the ins, out, and quirks of the framework and development process.

Accomplishments that we're proud of

VCT Valiance has allowed us to expand our knowledge on Kotlin's interoperability with JavaScript along with large-scale data parsing. We built the interop between backend JS function calls to be abstracted and error-safe, while also being easily scalable and modular. We were also able to develop a custom set of tools using C++ and Kotlin for parsing the large data sets in to more straightforward data for model training.

What we learned

We learned a lot about project planning and expectations throughout this hackathon. The data set's size proved challenging, and certainly warranted a more careful examination prior to starting development.

What's next for VCT Valiance

We hope to continue to develop VCT Valiance into a more capable and knowledgeable tool. Expanding it to support more in-depth analysis of pro games and other insights would prove challenging, but extremely helpful for the community!

Built With

Share this project:

Updates