Offline Adventures - lets start journey with AI
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built Offline Adventures is a local-first activity recommender designed to make choosing an activity quick, then get out of the way. The user selects available time, energy level, current weather, accessible places, group type, and interests. The app returns one primary adventure and two alternatives.…
Offline Adventures is an activity recommender that helps users choose an outdoor or indoor activity based on their preferences and current conditions. The app considers available time, energy level, weather, accessible places, group type, and interests to provide a primary recommendation and two alternatives. Each recommendation comes with a short explanation, mission, practical notes, a memory prompt, and a personalized twist.
The app uses a hybrid architecture with deterministic Python code for hard constraints and a local open-weight Llama3.1:8b-instruct-q4_K_M model from Ollama for semantic ranking and personalized output. The model is instructed not to invent activity IDs, locations, equipment, health advice, risky behavior, or laptop use during activities. This design ensures the app can operate locally without a hosted AI key or account, making it accessible and private.
The catalog includes 50 activities divided into outdoor (30), sheltered (8), and indoor-bridge (12) categories. The app was tested with various scenarios, including dry weather, low energy, photography and nature, and heavy rain with family, games, and nature interests. The app successfully generated suitable outdoor recommendations and sheltered or indoor alternatives when appropriate. The project was entered into the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass category.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
