A/B Testing Embedding Models (Helper Tool)
A helper step while building a bigger app: figuring out which embedding model fits a given content and language best.
About the project
When I was building a bigger app with vector search, I needed to know which embedding model fits that type of content and language best.
Instead of guessing, I made a small tool that let me compare them side by side.
A one-off helper step, not a standalone product. It did its job, helped me choose, and went on the shelf, exactly how these vibe-coded little utilities work.
Why I built it
- Built so I could pick the right embedding model.
- I compared the models side by side, not by feel.
- It did its job and went on the shelf, a one-off.
How it works (under the hood)
- A proxy server for Ollama embeddings.
- A simple HTML interface for A/B comparison.
- Node.js + Express, built quickly.
Built with
Want to build like this too?
I teach building apps without programming in the AI First course.
More from Tool
Shoutbase: Testimonials Manager
An internal tool for managing testimonials: a queue of who to ask, a collection of what we've gathered, and an overview of where we use them.
Toucan: A Shared Second Brain for the Team
A shared cloud knowledge base that the team and its AI agents draw from, a "second brain" connected to my local context1.
Running a Project's Marketing Through Claude Code
A marketing team as a set of Claude Code skills: a director, department managers, and specialists that run a project alongside me.