Vector Podcast website: https://vectorpodcast.comHaystack US 2025: https://haystackconf.com/2025/Federated search, Keyword & Neural Search, ML Optimisation, Pros and Cons of Hybrid searchIt is fascinating and funny how things develop, but also turn around. In 2022-23 everyone was buzzing about hybrid search. In 2024 the conversation shifted to RAG, RAG, RAG. And now we are in 2025 and back to hybrid search - on a different level: finally there are strides and contributions towards making hybrid search parameters learnt with ML. How cool is that?Design: Saurabh Rai, https://www.linkedin.com/in/srbhr/The design of this episode is inspired by a scene in Blade Runner 2049. There's a clear path leading towards where people want to go to, yet they're searching for something.00:00 Intro00:54 Eric's intro and Daniel's background02:50 Importance of Hybrid search: Daniel's take07:26 Eric's take10:57 Dmitry's take11:41 Eric's predictions13:47 Doug's blog on RRF is not enough16:18 How to not fall short of the blind picking in RRF: score normalization, combinations and weights25:03 The role of query understanding: feature groups35:11 Lesson 1 from Daniel: Simple models might be all you need36:30 Lesson 2: query features might be all you need38:30 Reasoning capabilities in search40:02 Question from Eric: how is this different from Learning To Rank?42:46 Carrying the past in Learning To Rank / any rank44:21 Demo!51:52 How to consume this in OpenSearch55:15 What's next58:44 Haystack US 2025

https://www.vectorpodcast.com/I had fun interacting with NotebookLM - mostly for self-educational purposes. I think this tool can help by bringing an additional perspective over a textual content. It ties to what RAG (Retrieval Augmented Generation) can do to content generation in another modality. In this case, text is used to augment the generation of a podcast episode. This episode is based on my blog post: https://dmitry-kan.medium.com/the-rise-fall-and-future-of-vector-databases-how-to-pick-the-one-that-lasts-6b9fbb43bbbeTime codes:00:00 Intro to the topic1:11 Dmitry's knowledge in the space1:54 Unpacking the Rise & Fall idea3:14 How attention got back to Vector DBs for a bit4:18 Getting practical: Dmitry's guide for choosing the right Vector Database4:39 FAISS5:34 What if you need fine-grained keyword search? Look at Apache Lucene-based engines6:41 Exception to the rule: Late-interaction models8:30 Latency and QPS: GSI APU, Vespa, Hyperspace9:28 Strategic approach9:55 Cloud solutions: CosmosDB, Vertex AI, Pinecone, Weaviate Cloud10:14 Community voice: pgvector10:48 Picture of the fascinating future of the field12:23 Question to the audience12:44 Taking a step back: key points13:45 Don't get caught up in trendy shiny new tech

John Berryman, founder of Arcturus Labs and co-author of "Prompt Engineering for LLMs," shares his journey from search technology to AI. He discusses the evolution of code search and how GitHub Copilot reshapes programming with AI. Berryman emphasizes the balance between automation and human oversight in coding practices. The conversation dives into retrieval-augmented generation (RAG) and the future of interactive digital artifacts, opening discussions on reliability and accountability in AI-generated content. His insights are both informative and thought-provoking.

In this intriguing conversation, Leo Boytsov, a Senior Research Scientist at AWS AI Labs and expert in vector search algorithms, shares enlightening insights from the cutting edge of search technology. He discusses the evolution of retrieval algorithms, challenges with large document handling, and how non-metric spaces can enhance similarity representation. Leo also reveals the potential of combining traditional and modern search methodologies, and the serendipitous discoveries shaping new industries in AI. A must-listen for tech enthusiasts!

Alessandro's talk on Hybrid Search with Apache Solr Reciprocal Rank Fusion: https://www.youtube.com/watch?v=8x2cbT5CCEM&list=PLq-odUc2x7i8jHpa6PHGzmxfAPEz-c-on&index=500:00 Intro00:50 Alessandro's take on the bbuzz'24 conference01:25 What and value of hybrid search04:55 Explainability of vector search results to users09:27 Explainability of vector search results to search engineers13:12 State of hybrid search in Apache Solr14:32 What's in Reciprocal Rank Fusion beyond round-robin?18:30 Open source for LLMs22:48 How we should approach this issue in business and research26:12 How to maintain the status of an open-source LLM / system 30:06 Prompt engineering (hope and determinism)34:03 DSpy35:16 What's next in Solr

Video: https://youtu.be/dVIPBxHJ1kQ00:00 Intro00:15 Greets for Sonam01:02 Importance of metric learning3:37 Sonam's background: Rasa, Qdrant4:31 What's EmbedAnything5:52 What a user gets8:48 Do I need to know Rust?10:18 Call-out to the community10:35 Multimodality12:32 How to evaluate quality of LLM-based systems16:38 QA for multimodal use cases18:17 Place for a human in the LLM craze19:00 Use cases for EmbedAnything20:54 Closing theme (a longer one - enjoy!)Show notes:- GitHub: https://github.com/StarlightSearch/EmbedAnything- HuggingFace Candle: https://github.com/huggingface/candle- Sonam's talk on Berlin Buzzwords 2024: https://www.youtube.com/watch?v=YfR3kuSo-XQ- Removing GIL from Python: https://peps.python.org/pep-0703- Blind pairs in CLIP: https://arxiv.org/abs/2401.06209- Dark matter of intelligence: https://ai.meta.com/blog/self-supervised-learning-the-dark-matter-of-intelligence/- Rasa chatbots: https://github.com/RasaHQ/rasa- Prometheus: https://github.com/prometheus-eval/prometheus-eval- Dino: https://github.com/facebookresearch/dino

00:00 Intro00:30 Greets for Doug01:46 Apache Solr and stuff03:08 Hello LTR project04:42 Secret sauce of Doug's continuous blogging08:50 SearchArray13:22 Running complex ML experiments17:29 Efficient search orgs22:58 Writing a book on search and AIShow notes:- Doug's talk on Learning To Rank at Reddit delivered at the Berlin Buzzwords 2024 conference: https://www.youtube.com/watch?v=gUtF1gyHsSM- Hello LTR: https://github.com/o19s/hello-ltr- Lexical search for pandas with SearchArray: https://github.com/softwaredoug/searcharray- https://softwaredoug.com/- What AI Engineers Should Know about Search: https://softwaredoug.com/blog/2024/06/25/what-ai-engineers-need-to-know-search- AI Powered Search: https://www.manning.com/books/ai-powered-search- Quepid: https://github.com/o19s/quepid- Branching out in your ML / search experiments: https://dvc.org/doc/use-cases- Doug on Twitter: https://x.com/softwaredoug- Doug on LinkedIn: https://www.linkedin.com/in/softwaredoug/

00:00 Intro00:21 Guest Introduction: Eric Pugh03:00 Eric's story in search and the evolution of search technology7:27 Quepid: Improving Search Relevancy10:08 When to use Quepid14:53 Flash back to Apache Solr 1.4 and the book (of which Eric is one author)17:49 Quepid Demo and Future Enhancements23:57 Real-Time Query Doc Pairs with WebSockets24:16 Integrating Quepid with Search Engines25:57 Introducing LLM-Based Judgments28:05 Scaling Up Judgments with AI28:48 Data Science Notebooks in Quepid33:23 Custom Scoring in Quepid39:23 API and Developer Tools42:17 The Future of Search and Personal ReflectionsShow notes:- Hosted Quepid: https://app.quepid.com/- Ragas: Evaluation framework for your Retrieval Augmented Generation (RAG) pipelines https://github.com/explodinggradients...- Why Quepid: https://quepid.com/why-quepid/- Quepid on Github: https://github.com/o19s/quepid

00:00 Intro01:54 Reflection on the past year in AI08:08 Reader LLM (and RAG)12:36 Does it need fine-tuning to a domain?14:20 How LLMs can lie17:32 What if data isn't perfect21:21 SWIRL's secret sauce with Reader LLM23:55 Feedback loop26:14 Some surprising client perspective31:17 How Gen AI can change communication interfaces34:11 Call-out to the Community

Louis Brandy, VP of Engineering at Rockset, talks about using SQL for Vector Search. They discuss Rockset's product, AI capabilities like ANN index, hybrid search, and future predictions. Louis emphasizes the importance of vector databases and the challenges in optimizing search engines with vector support.