Skip to main content
usedby
Search tools, companies

How Airtable uses Zilliz

As published on zilliz.com. Checked by usedby on Oct 17, 2026.

What happened

Airtable's Data Infrastructure team runs self-hosted Milvus as the vector database behind its semantic search. It stores row embeddings for each customer base and retrieves the most relevant rows as context for the Omni AI chat and for linked record recommendations.

Summary written by usedby from the source page, in English. The figures are those of Zilliz and Airtable, not ours.

Milvus emerged as the best fit: its distributed nature supports massive multi-tenancy and allows us to scale ingestion, indexing, and query execution independently, delivering performance while keeping costs predictable.

From the page. zilliz.com

Zilliz gives product teams a strong retrieval foundation for building AI-native workflows on structured and unstructured data.

Aria Malkani, Infra Software Engineer, Airtable. Source, captured Oct 6, 2026

What the story claims, and what we checked

What we compared with the page.

  • The passage quoted aboveCheckedCopied word for word from the page, near the name of Airtable.
  • Airtable uses ZillizCheckedConfirmed line.
  • The result itselfNot checkedWe quote it; we did not measure it.

Same company, same tool or same industry.

1-2 weeks“Shipped sandboxed code execution in 1-2 weeks, roughly 10x faster than building it in-house”ClickUp uses E2B. Same industry: DevTools & Infrastructure1B+“1B+ Simultaneous viewers”Miris uses CoreWeave. Same industry: DevTools & Infrastructure50%of production time savedMiro uses Runway. Same industry: DevTools & Infrastructure