One of the reasons why I host Vector Podcast is the unique opportunity to meet the makers in the Search space and learn from them. My conversation with Atita Arora, Search Relevance Consultant at OpenSource Connections, gave me both the chance to learn new things, and enjoy the drive, energy and dedication to the field of search engine development, that Atita is full of. It is not a typical episode, because we spoke about life, profession, moving around, open source, women of search, but also vector search for #ecommerce. As usual, you will find quite a bit of self-study material on the topic, including code, datasets and blogs. Be sure to check the show notes.

While working on this blog post I had a privilege of interacting with all neural search frameworks key developers / leadership: Malte Piestch (Haystack), Amin Ahmad (Vectara), Florian Hönicke (Jina AI), David Mezzetti (txtai), Daniel Vassilev (Relevance AI), George Sivulka and Swetha Revanur (Hebbia), Aarne Talman and myself (Muves). The…

Around two weeks ago I had a big honor to deliver the keynote at the Haystack EU conference held in Berlin, Germany. In it, I’ve shared some of my learnings from interviewing Vector Search makers in the Vector Podcast. Worth sharing them here: Malte Pietsch (CTO Deepset, Haystack): Metric blindness …

This year Berlin Buzzwords conference has reopened to offline participants for the first time after pandemic started. The conference has adopted the hybrid mode, with all talks streaming online as well. These were 2 active days full of technical talks in 3 main trains: Search, Store and Scale. I also…

Last week Thursday I spoke at Pinecone’s workshop by invitation of Greg Kogan, VP Marketing, Pinecone — one of the 7 vector databases described in my blog. The main topic was around how to choose a vector database, but we also discussed what is vector search, its use cases, aspects…

In July I got an invitation from the Sease team to present at the London IR Meetup, where I spoke about players, algorithms, software and use cases in Vector Search. In this presentation I’ve used the opportunity to give an update on the first season of Vector Podcast: it…

Note: this video presentation is based on my survey of Vector Databases. Last week at the NLP Meetup organized by deepset I’ve presented on the Landscape of Vector Databases and neural search frameworks. Max Irwin was my co-presentor to deep-dive into the BuddyPQ algorithm we have invented together as part of Sisu Team during Billion-Scale Approximate Nearest Neighbour Search Challenge. The main achievement of our algorithm is 12% higher recall over baseline FAISS algorithm.

Vector search and databases is a buzzing and an emerging space. I’m happy to announce the launch of my podcast to talk with the creators and makers of this new search wave. Give it a listen and let me know what you think. Apple Podcasts: https://podcasts.apple.com/us/podcast/greg-kogan-pinecone-on-the-emerging-field-of/id1587568733?i=1000536632146 YouTube: https://www.youtube.com/c/VectorPodcast Spotify:

A detailed comparison of Milvus, Pinecone, Vespa, Weaviate, Vald, GSI and Qdrant — While working on this blog post I had a privilege of interacting with all search engine key developers / leadership: Bob van Luijt and Etienne Dilocker (Weaviate), Greg Kogan (Pinecone), Pat Lasserre, George Williams (GSI Technologies Inc), Filip Haltmayer (Milvus), Jo Kristian Bergum (Vespa), Kiichiro Yukawa (Vald) and Andre Zayarni…

This blog goes into detail about the work we did as Team Sisu for the Billion-Scale Approximate Nearest Neighbor Challenge: https://big-ann-benchmarks.com/index.html Long story short: we have improved recall by 12% on 10M dataset over baseline FAISS. Update: We have presented part of this work in the NLP meetup by deepset:

This year’s Berlin Buzzwords was particularly focusing on what can become the future of search — vector search, amongst other really cool topics, like scaling Kafka, distributed systems tracing with Opentelemetry and increasing the job satisfaction. …

It is exciting to read the latest research advances in the computational linguistics. In particular, the better language models we build, the more accurate downstream NLP systems we can design. Update: if you are looking to run neural search with latest Solr versions(starting version 8.x), I have just published a…

After publishing the blog post on neural search with BERT and Solr (6.6.0), I got a few questions on how to run this with version 8.6.x of Solr. It took me a few days of going back and forth, and quite honestly a bit of despair, and finally a helping…

Neural Search in Elasticsearch: from vanilla to KNN to hardware acceleration — In two previous blog posts on my journey with BERT: Neural Search with BERT and Solr and Fun with Apache Lucene and BERT I’ve taken you through the practice of what it takes to enable semantic search powered by BERT in Solr (in fact, you can plug in any other…