Built a weather app with AI-generated summaries and city bookmarking—feedback welcome (took me 9-10 days; learned a lot)

A developer built a weather app featuring AI-generated natural language summaries, dynamic backgrounds, city search with live suggestions, and bookmarking with localStorage. The project, built with React and Vite and deployed on Vercel, took 9-10 days and involved challenges such as debouncing API calls, handling duplicate city names, and persisting bookmarks.

Hey everyone, wanted to share a project I just finished. It's a weather app, but I tried to push past the typical "fetch API, show temperature" tutorial project. Features: AI-generated natural language weather summaries instead of raw numbers Dynamic backgrounds that change based on real weather conditions City search with live suggestions as you type Bookmark/favorite cities, persisted with localStorage Stack: React + Vite, deployed on Vercel Stuff that actually gave me trouble: Debouncing the city search so it doesn't spam the API on every keystroke Duplicate city names across countries messing up search results Bookmarks kept vanishing on refresh until I realized I wasn't persisting state properly Syncing background image changes with API response timing without a jarring flash It took about 9-10 days total, mostly because I kept refactoring state management once bookmarks, search, and multi-city display all had to interact. Still learning, so genuinely open to criticism — especially on component structure, since that's where I feel least confident.