We have all been there. You need a simple background remover or want to compare DALL-E 3 pricing against Midjourney, but search results are clogged with heavily optimized SEO spam, aggressive paywalls, and outdated blog posts.
I got tired of clicking through dozens of spammy sites just to find a simple, functional free stock photo resource or a reliable AI image upscaler. So, I did what any developer does: I spent a weekend building my own curated database hub, ImageNav, to solve this tool-fatigue problem once and for all.
Here is how I engineered a lightning-fast, zero-database catalog of 85+ AI image tools that loads in milliseconds and costs nothing to host.
Why Databases for Small Catalogs is an Anti-Pattern
When building directory sites, the knee-jerk reaction is to spin up a PostgreSQL instance or hook up a heavy headless CMS. But for a dataset of ~100 curated tools, this adds unnecessary latency, architectural overhead, and database query costs.
Instead, I opted for a static-first architecture using Next.js and structured JSON files.
Every tool on ImageNav is represented as a typed object in a TypeScript schema. By storing the data locally:
- Performance: Pages are compiled statically at build time. Search and filtering happen entirely client-side, making the UX feel instantaneous.
- Cost: The site is hosted on Vercel with zero database cold-start issues, completely within the free tier.
- Maintainability: Data updates are as simple as editing a local JSON file or merging a quick Git pull request.
High-Performance Client-Side Filtering
To make sorting through 85+ tools seamless, I built a lightweight React Hook that manages multi-select facets (Categories, Pricing Models, and Features).
Instead of pulling in heavy third-party search libraries, I combined a lightweight regex fuzzy search with Native JS array filtering. Since the dataset size is highly optimized, these array operations execute in under 2ms even on low-end mobile devices, keeping the UI highly responsive.
Automating Link & Integrity Checks
One of the biggest issues with curated directories is link rot and outdated pricing. To tackle this, I wrote a Node.js script that runs weekly via GitHub Actions. It sends lightweight HEAD requests to all cataloged URLs to verify status codes and flags any tool that goes offline or changes redirects.
ImageNav is now live at https://imagenav.net. If you are looking for a clean, ad-free way to find your next design or AI generator tool, check it out and let me know what you think of the UX!
Top comments (0)