INSIGHT_PROTOCOL

I unfortunately am not able to add a video, however the app is incredibly simple and easy to use. just make sure to use the ML_EXPERIMENTATION branch link here: (https://github.com/PhantomPilot4729/C9_Automated_Scout.git)

Inspiration

I was inspired by Cyberpunk 2077 for the UI, as well as my discovery of the access for open source AI models

What it does

what it does, is it takes real-time match data (or manual input data) and gives 2 types of advice based off of that. the first type is quick, non-ai advice to get a quick overview of the strategy, especially econ advice. then, it gives AI coaching advice, to provide new and creative strategies to win.

How we built it

I'm going to be so deadass, i vibe coded the whole thing. i have no idea how to code in C#, but i knew what i wanted and i knew how to get there, so all i had to do from there was get Junie to do all the heavy lifting

Challenges we ran into

A lot of the challenges i ran into revolved around getting the code to communicate with the locally-stored AI model. also, trying to get the AI to provide a visual representation of where various players and util should go, to be added in a later iteration

Accomplishments that we're proud of

I'm really proud of how i was able to go from knowing nothing about how AI actually worked, and how to actually train AI, to having done that in a really short time period. I also think that my product will eventually give really good coaching advice, with more training data.

What we learned

see above, generation of visual images sucked, and getting various programs to communicate with each other was... difficult to say the least

What's next for INSIGHT_PROTOCOL

What's next is adding the visual of the map with pins for locations, routes, and util placements, as well as training the AI more and who knows, maybe train its tactical advice based off special forces training

Built With

• .xaml
• c#
• grid
• junie
• ollama