Sprinkle: Finding a Better Home for the Things We Donate
This is a submission for Weekend Challenge: Generosity Edition . What I Built When I was younger, my family went through times when we needed help from local charities, food pantries, clothing programs, and things like that. I do not need to get into all the details, but I remember being surprised by how generous people could be when we really needed it. That stuck with me, and it is a big part…
This story is about Sprinkle, a web application designed to simplify the process of donating useful items to local organizations that need them. The creator of Sprinkle, who grew up needing help from charities and food pantries, wanted to find an easier way to donate items like coats, food, hygiene products, electronics, and household items.
Sprinkle takes in a description of the items someone wants to donate in plain language, along with a ZIP code or city. The app then searches for nearby organizations, reads their donation and needs pages, and matches the items with the suitable organizations. Each result is verified with a label and the sources used, allowing users to double-check the information before donating.
The app uses Google's Gemini AI to handle the complex tasks of parsing donation descriptions, discovering nearby organizations, researching their donation policies, and matching the items. The code for Sprinkle is available on GitHub, and the app is built with React, TypeScript, and runs through a single Cloudflare Worker. The Gemini API key is kept on the server, and the research is split into four parts: parsing donation descriptions, discovering nearby organizations, researching donation policies, and matching the items.
The verification system is crucial to ensure the app's accuracy. Results are labeled as Confirmed, Likely, Call first, or Unknown, depending on the confidence level and the availability of a valid source. The UI then displays progress updates through server-sent events, providing transparency on what the app is currently doing. Sprinkle is licensed under Apache 2.0, and it does not require users to create an account or provide any personal information.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
