StratLens – AI Assistant Coach

See the Game Beneath the Game.

Comment

Inspiration

Esports teams collect massive amounts of data, yet coaches still spend hours reviewing VODs trying to answer a simple question: why did we actually lose? Inspired by Moneyball, we wanted to move beyond raw stats and surface the hidden causal links between individual player decisions and team-wide outcomes. StratLens was born to help coaches see the game beneath the game.

What it does

StratLens is an AI-powered Assistant Coach that connects micro-level player mistakes—such as isolated deaths, mistimed rotations, or poor positioning—to macro-level strategic consequences like lost map control, failed executes, and round losses.

Instead of dashboards full of numbers, StratLens delivers clear, explainable coaching insights, role-specific recommendations, and impact scores that show what to fix first and why it matters.

How we built it

We built StratLens as a modular web application using:

  • GRID esports match data for structured historical analysis
  • A custom analytics engine to detect recurring micro mistakes
  • A causal impact layer to map mistakes to macro outcomes
  • AI-driven insight generation, assisted by Junie, to translate analytics into natural-language coaching insights
  • JetBrains IDEs to rapidly prototype, refactor, and scale the system architecture

The system was designed MVP-first, with a clean path to real-time analysis and advanced features.

Challenges we ran into

  • Translating complex match data into coach-friendly insights
  • Avoiding black-box ML and ensuring explainability
  • Scoping the project to stay hackathon-realistic while showing production potential
  • Designing micro → macro mappings without oversimplifying gameplay nuance

Accomplishments that we're proud of

  • Built a working Moneyball-style causal analysis for esports
  • Delivered clear, actionable insights instead of raw statistics
  • Designed a system that scales to real-time analysis and multi-game support
  • Created a coach-focused product that feels practical, not experimental

What we learned

  • Coaches value clarity over complexity
  • Explainable AI builds more trust than predictive black boxes
  • Strong storytelling is just as important as strong analytics
  • Designing for real users keeps technical ambition grounded

What's next for StratLens – AI Assistant Coach

  • Real-time in-game analysis for timeouts and pauses
  • “What-if” simulations to quantify how corrected mistakes change outcomes
  • Auto-generated practice drills linked directly to detected weaknesses
  • Role-specific coaching plans for long-term player development
  • Multi-game support across VALORANT, League of Legends, and beyond

StratLens aims to become a complete data-to-training coaching loop for modern esports teams.

Built With

Share this project:

Updates