Software Engineering Daily Turbopuffer with Simon Hørup Eskildsen
39 snips
Sep 30, 2025 
Simon Hørup Eskildsen, co-founder of TurboPuffer and former Shopify infrastructure engineer, joins security expert Gregor Vand to discuss the fascinating world of vector databases. They dive into how TurboPuffer was shaped by early AI experiments and the challenge of storage costs. Simon explains its design for lightning-fast queries and the importance of unique indexing strategies. He shares success stories from companies like Cursor and Notion, emphasizes a focus on commercial clients, and hints at exciting features in the pipeline, all while sporting a fun pixel-art aesthetic.
AI Snips
Chapters
Transcript
Episode notes
From Shopify Tuning To Founding TurboPuffer
- Simon left Shopify after a decade and spent two years helping startups with infrastructure before founding TurboPuffer.
- He started hacking TurboPuffer in summer 2023 after realizing vector storage economics were broken for many apps.
Vector Storage Amplifies Data Size
- Vector embeddings amplify storage size dramatically, often 20–30x compared to raw text.
- That amplification plus in-memory storage made early vector databases prohibitively expensive for many products.
New Hardware And APIs Made It Possible
- NVMe SSDs and object storage consistency enabled a new storage architecture that is far cheaper than DRAM.
- Compare-and-swap on object storage was the final primitive needed to build TurboPuffer's design.
