AI-Enhanced Board Game — Engineering Play with Intelligence

The Spark

It started as a class group project and then led to a question: Could a board game learn how you play and respond proactively?

Between my long interest in mechanical design and a general knowledge of neural networks, the idea found my credibility in my analysis. I thought 'EASY' - just merge tactile gameplay with adaptive intelligence. Instead of static rules, what if each match evolved, learning from how players strategize, take risks, and bluff? Similar to a digital chess game player, but integrated into a board game.

That question became my first serious attempt to build a physical-digital hybrid system, an AI-enhanced board game that learns from experience, like a human opponent would. How much traction would it get in the $13.95 B market with an annual 8.7% compound annual growth rate?

Building Intelligence into the Board

The prototype’s brain is a Raspberry Pi 4 running Python scripts that collect player-move data through capacitive sensors embedded under the tiles. Bronson, my teammate, must be given credit here as his programming skill helped me through a big hurdle of how each tile communicates its state to the main board controller via I²C and GPIO, translating every move into a structured dataset.

We approached it by creating a lightweight neural-network model built in TensorFlow Lite, which evaluates this incoming data from I²C and GPIO in real-time. Its job is to detect patterns — aggression, hesitation, cooperation — and adapt its responses accordingly. We called it Mr. General. The result took a minute to be accurate and more predictable from a programmer's perspective, and finally, we got a reactive gameplay that showed tangible results.

The AI adjusts difficulty dynamically, much like reinforcement learning in robotics, though we hardcoded this part with if-else statements to avoid complications in logic. We later thought of an idea of adding randomized noise and reward-based weighting to make the system more human and less deterministic, but we could not get to it.

Vision Boards, Collages, and Context

Before I wrote a single line of code, I designed the experience so that my team would also be as hyped as me. I built vision boards that mapped my vision with sample logic of emotional states to visual cues — warmth for progress, blue for logic, red for risk. Collages of dreams, examples, the possibilities.

My teammate Griffin put forward the thought of integrating a hexagonal tile pattern on the board, the reasoning was that it was most efficient for motion detection while keeping aesthetic symmetry. My thought - Hexagons = "Cool".

Hardware Meets Software

The design employed 3D-printed tile casings, modeled in Fusion 360, with possible conductive paint circuits to eliminate traditional wiring clutter.

The AI logic runs parallel to a microcontroller (ESP32) handling real-time inputs to minimize latency. We later ideated that a communication bridge using MQTT over local Wi-Fi enables scalable multi-board communication so that multiplayer expansion is possible.

Challenges and Breakthroughs

1. Latency vs. Intelligence: My first iteration ran entirely on Python, causing noticeable lag. I ported the inference model to TensorFlow Lite C++ for faster computation. Bronson contributed heavily to complement my skill gap.

2. Data Representation: Translating human intuition into numbers was tricky. We created a custom scoring metric combining entropy (for unpredictability) and consistency (for skill) using Gemini as our coder.

3. User Feedback: Adding subtle audio cues increases immersion — the AI “reacts” with tone rather than text on Pi, making the board feel alive.

   
   

Engineering as Storytelling

The engineering design became a form of storytelling, where AI served as the co-author rather than the referee. Building this project taught me that creativity and computation aren’t opposites; they’re collaborators. Vision boards guided the algorithms; circuit schematics shaped emotion.

The Takeaway

This project pushed me to think about design systems holistically - how hardware, software, and psychology can converge to create experiences that feel both engineered and alive.

The AI-Enhanced Board Game fosters play that evolves. It’s about designing intelligence that entertains, learns, and surprises, a true collaboration between human intuition and machine logic.