Top all-time · #13949
TopK
Company blog · topk.io/blogaidata
TopK is the company blog at topk.io, writing about ai and data. Since 2025, 5 different Hacker News readers have linked it 7 times. No feed was found, so read it on the site itself.
Integrated inference and storage. Flexible retrieval primitives for AI teams pushing the frontier of accuracy, performance, and scale.
What Hacker News says?
> The problem with a vector primary index We've realized this a long time ago at TopK and built a flexible serverless search engine from scratch. Supports dense/sparse vectors, late interaction, lexical search, indexed regex, filtering, and custom scoring in one query. - https://www.topk.io/blog/vector-dbs-are-the-wrong-abstractio... - https://www.topk.io/blog/topk-embed-v1
Comment linking /blog/topk-embed-v1
on “RIP, vector database”Submitted Why Vector DBs Are the Wrong Abstraction – and What We Built Instead as a story · 23 points
HN discussionSubmitted Real-world vector DB performance across the most popular providers as a story · 12 points
HN discussionSubmitted Binary Vector Search at 350GB/S Using ARM Neon as a story · 7 points
HN discussionSubmitted SMVE: Multi-Vector Retrieval That Just Works as a story · 1 points
HN discussionSubmitted Beyond RRF: Improving Hybrid Search by Up to 7.8% as a story · 1 points
HN discussion