Golden Hour Catcher
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built Golden Hour Walk tells you exactly when to leave the house to catch the best light of the day. You give it your location (or press "Use my location"), a date, and whether you want a morning or evening walk. You also say how long you want to walk and how far the spot is. It works out when golden…
Golden Hour Catcher is an open-source application that helps users determine the best time to start a walk outside to capture the ideal lighting of the day. By providing your location, desired date, and walk details, the app calculates golden hour start and end times, sunrise and sunset, and your departure time. The result is a concise, friendly plan like "Leave at 5:40 PM so you reach the water as golden hour starts."
The primary motivation behind Golden Hour Catcher is to encourage people to step outside, as many skip walks out of disinterest. Golden hour offers a specific time to leave and an ever-changing sky, making it perfect for those who spend their days in front of a screen. The application is a single index.html file without a backend, which means it deploys as a static site.
The core idea behind the app is simple: code handles the facts, while the model generates the sentences. The sun times are derived from a solar calculation performed in the browser using JavaScript, ported from Python's astral library. The wording for the plans is provided by an open-weight model running inside the user's browser using WebLLM and WebGPU.
The default model is Qwen 2.5 1.5B, with options for Llama 3.2 1B and 3B available in a dropdown. The model only receives a list of facts and produces two sentences, ensuring the output remains accurate. A guardrail in place checks the model's suggested times against the calculated facts before presenting them to the user. This way, the app eliminates the risk of displaying incorrect times and maintains the integrity of the information provided.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
