Game AI Opponent for Nine Men’s Morris

Game AI Opponent for Nine Men’s Morris

This is a focused engineering case study from my Nine Men’s Morris Flutter game, which is also listed on this profile. I built a computer opponent that chooses legal moves across placement, movement, flying, and capture phases. It uses minimax search with alpha-beta pruning, phase-aware board evaluation, and deterministic tie-breaking.

The opponent plugs into the same immutable rules engine as local two-player play. Search depth maps to Easy, Medium, and Hard settings; the game UI handles the AI turn and a separate capture decision while guarding against stale results after a restart or state change.

My contribution: game state modeling, legal-action API, search and evaluation, difficulty integration, and the Flutter play flow. Stack: Flutter, Dart, minimax, alpha-beta pruning.

This is classical game AI, not an LLM or generative AI feature. It is a deep dive into the AI opponent within the existing game, not a separate client engagement. The game remains in development. The cover is a conceptual illustration, not an app screenshot.