AI productsModelsAgentsData platformsInfrastructure
Hackathon project
LLM-powered digital assistant for VCT Hackathon
An AI-powered scouting assistant that analyzes VALORANT player data and recommends team compositions based on roles, performance, and synergy.
View the project on Devpost ↗The problem
Team managers and coaches need to compare player performance and team fit, but manually analyzing metrics is slow and prone to inconsistency.
What we built
01
Built an LLM-powered assistant for questions about VALORANT esports players and team composition.
02
Connected AWS Bedrock Agents to a structured player-performance knowledge base stored in Amazon S3.
03
Used the most recent 90 days of player data to support scouting and role-aware recommendations.
04
Shipped a Next.js frontend, Django backend, AWS EC2 API layer, and AWS Amplify deployment.
Capabilities
- Player performance analysis
- Role-aware team composition
- LLM question answering
- Scouting workflow support
- Cloud AI integration
Technology
AWS Bedrock AgentsAmazon S3Amazon EC2AWS AmplifyDjangoNext.js
What changed
- A faster way to explore player performance data
- Team recommendations that consider roles and synergy
- A working cloud architecture for AI-assisted esports scouting
Have a difficult system to build?
Let’s turn the hard part into something useful.