Touch grass, and touch glass on a padel court
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built Arranging a padel match can be surprisingly tedious. You need four players of roughly similar ability, available at the same time, preferably with compatible expectations. In practice, this means checking club apps, messaging WhatsApp groups, negotiating times, and chasing people for confirmation.…
Padelpot is an open-source experiment that uses AI agents to simplify the process of arranging a doubles padel match. Each agent represents a player and learns their preferences, while negotiating with other agents to find a suitable time and location. Their interactions are limited to basic information, with privacy and commitment rules enforced by the application.
In a demo, four fictional players - Alba, Nico, Luz, and Teo - use a chat interface to negotiate a Thursday evening match. Their agents resolve a disagreement about the start time and obtain approval from all four players. The Java application, built with The Pipeline Framework, utilizes open-weight Gemma models for interviews and negotiations.
It includes setup instructions, synthetic player profiles, automated tests, and reproducible demonstration scenarios. The code is available on GitHub and the model inference can be done locally using Ollama or through OpenRouter. The application handles matchmaking rules, privacy policies, and final approvals, ensuring that no more than four players are confirmed for a match at any given time.
The project emphasizes the importance of openness in open innovation, allowing for local experimentation with open-weight models, inspection of results, and extension of the execution model and application architecture.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
