Dima Kan

Background

Dima Kan is Product Director of Search at Aiven, a managed database platform. He is the founder and host of Vector Podcast, launched in October 2021 to interview the creators of the emerging vector-search wave.

He spotted that wave early: days before the podcast started he published “Not All Vector Databases Are Made Equal” on Medium — a side-by-side of Milvus, Pinecone, Vespa, Weaviate, Vald, GSI and Qdrant written with people from each of those teams, which he reports has been read over 120K times.

He also authored the article introducing Wormhole Vectors and their implementation in OpenSearch, applying Trey Grainger’s research to production architecture.

Media

• Vector Podcast — founder and host; https://www.vectorpodcast.com/ (YouTube)
• “Not All Vector Databases Are Made Equal” (Medium, October 2021) — early comparative survey of the vector database landscape

Articles in This Vault

• Beyond Hybrid Search - Traversing Vector Spaces with Wormhole Vectors

• Wormhole Vectors Beyond Hybrid Search in OpenSearch

Key Contributions

• Practical implementation guide for Wormhole Vectors in OpenSearch
• Production architecture bridging lexical, dense, and behavioral vector spaces

Concepts

• Wormhole Vectors
• Hybrid Search
• Semantic Search