Case Study: Cursor achieves 20x lower code retrieval costs and faster scaling with turbopuffer

Cursor switched from a vector database with a traditional storage architecture to turbopuffer, cutting costs by 20x and unlocked the ability to scale effortlessly as they grow

The customer, Cursor, faced significant scaling challenges and high costs with their previous vector database as they needed to manage an enormous number of vector indexes for code retrieval in their AI-powered code editor.

By switching to turbopuffer's serverless vector database, Cursor achieved a 20x cost reduction and unlocked effortless scaling, handling over 80 million namespaces and write peaks of 10GB/s. The implementation of turbopuffer also contributed to a 23.5% improvement in agent accuracy and drastically reduced codebase onboarding times from hours to seconds.

Cursor

Sualeh Asif

Co-founder and CTO