Etienne Dilocker on Vector Search Engines and Weaviate

Vector Search Engines are powering the the next generation of search. Instead of relying on things like BM25 or TF-IDF representations, we use neural representations to compare semantic similarity. Dot products to calculate the similarity between all vectors would be very time consuming. Hierarchical Navigable Small World Graphs (HNSW) are a cutting-edge new algorithm used by Weaviate to speed up this search. This podcast explores HNSW, Neurosymbolic search and filtering, and many more!

I hope you find this interesting, happy to answer any questions left on the YouTube video!

Check out this Introduction to Weaviate: Very well organized, no headaches getting started!

Chapters
0:00 Introduction
1:30 Why Vector Search Engines?
3:02 Hierarchical Navigable Small World Graphs and Scaling
9:00 Setup Questions
12:57 What is the compute heavy part?
17:15 Neurosymbolic Databases
23:53 Weaviate for Language Models
26:17 Weaviate for HuggingFace or PapersWithCode Datasets
30:43 Weaviate Modules
33:58 Text-to-Image Vector Search
36:54 What inspired Etienne to work on this?

[etiennedilocker]

Thanks for having me on the podcast. I really enjoyed our conversation and I'm happy Weaviate is helping so many users, be it in academia, business or just for fun.

[sonOfLiberty]

Wow I designed search engines by myself in my workplaces. I will enjoy this one for sure :) will listen to it later. Thank you so much <3

[tensorstrings]

36:08 let’s run with this idea some more - general multi-modal zero-shot model. What’s the arbitrary transform between any context for any?

[markadyash]

If you have added video conversion would be greaful