PlayNext

This project, called PlayNext, was created to help friends and fellow gamers overcome the time-consuming task of deciding what to play. The website simplifies the decision-making process by allowing users to select their desired mood, the amount of time they have, and the genre of game they prefer. A local open-weight AI model, Gemma, powered by Ollama, then generates three game recommendations based on the user's preferences.

The concept for PlayNext came about when the creator noticed that their friends often spent more time deliberating over game choices than actually playing. To address this issue, PlayNext was developed - an easy-to-use platform that streamlines the process of selecting a game.

The user experience is simple: users choose their desired mood (Bored, Relaxed, Competitive, or Adventurous), the amount of time they have (15 minutes, 30 minutes, or 1 hour+), and the type of game they prefer (Surprise Me, Action, Adventure, Puzzle, or Multiplayer). After submitting their choices, the script.js reads the selections and constructs a prompt, which is sent to Ollama's local API (http://localhost:11434/api/generate).

The Gemma model then generates three game recommendations in JSON format, which the webpage parses and displays in three interactive game cards.

PlayNext demonstrates the power of open innovation, as anyone can modify the AI model, adjust the prompt, or tailor the project to suit their own gaming preferences. The project code, along with a demo video and source code, is available on GitHub for anyone to explore, contribute to, or adapt for their own gaming needs.