Shaurya Verma

Project

AI Poker

Browser poker against software opponents with full table state.

A poker game against rule-based opponents: table state, turn logic, hand evaluation, and decisions under incomplete information.

Why I built it

Poker was a good excuse to build something where the software never has complete information. Unlike a puzzle solver, an opponent has to make a reasonable decision without knowing every card.

What it does

The browser game manages a full poker table, evaluates hands, advances turns, and lets a player compete against rule-based software opponents.

Why it matters

The interesting part was not the visual table; it was keeping every transition legal and making the opponents behave consistently from partial state. It strengthened how I think about state machines and game rules.

Under the hood

  • React drives the table and turn-state interface.
  • Hand evaluation converts card combinations into comparable ranks.
  • Rule-based opponents choose actions from visible table state rather than hidden player cards.