Show HN: Waiting for LLMs Suck – Give your user a game (github.com)

A new tool called "Waiting Arcade" has been introduced, designed to keep users engaged with mini-games while they wait for lengthy tasks, such as running large language models (LLMs) or uploads. This innovative solution provides a set of simple, one-button games—ranging from "Jellyfish Drift" to "Rhythm Tap"—that share a unified design framework, including achievements and scoring mechanisms. All games leverage minimal pixel art and are optimized for rapid deployment in web applications, assuring compatibility across various platforms. The significance of this tool for the AI/ML community lies in its ability to enhance user experience during wait times, which can often lead to frustration. The arcade's single-button interface simplifies interaction, making it accessible across devices and minimizing distractions. Technical details include SSR safety, the option for local storage to track high scores, and a lightweight framework that does not burden system resources. Developers can easily implement this component in their applications, with shared game features fostering a consistent user experience. This could lead to higher retention rates during waiting periods, making it a valuable addition for those integrating complex AI tasks into user interfaces.