How to use data to optimize AI apps

Today we’re sharing a video from one of our customers, Find AI. Last week alone, they made 19 million requests to OpenAI. They use Velvet to analyze, evaluate, and fine-tune their AI features.

What you’ll learn in this video:

• Run cost attribution and batch analysis

• Evaluate models using training sets

• Fine-tune gpt-4o-mini with historical data

Watch this in-depth tutorial from the founder of Find AI.

2 lines of code to get started

Velvet is a proxy to warehouse LLM requests to your database. Use the AI gateway to analyze and optimize AI features. Just 2 lines of code to get started!

Read the docs →

Free up to 10k requests / month

Use Velvet free up to 10k requests per month. Get started quickly with a hosted database, or connect your own PostgreSQL database.

Velvet pricing →