Introduction to Vector Search and Vector Databases with Weaviate - Part 2

Dive into a comprehensive 56-minute webinar exploring the fundamentals and applications of vector search technology. Master the mathematical principles behind similarity calculations and data retrieval while learning how vector search addresses the limitations of basic search at scale. Explore Approximate Nearest Neighbor (ANN) algorithms, with special focus on HNSW implementation and its performance optimization. Follow along with practical demonstrations using the Wikipedia dataset in Weaviate, and understand how CRUD operations integrate with traditional database features. Get hands-on experience implementing vector search in Python, working with Streamlit integration, and exploring multimodal and multilingual search capabilities. Learn about hybrid search techniques that combine keyword and vector approaches, and discover advanced applications with AI agents. The session concludes with an interactive Q&A addressing real-world implementation challenges and best practices.

Syllabus

0:07 Introduction to Vector Search
0:24 Overview of Vector Databases and Vector Search
2:26 Keyword Search Explanation
4:19 Introduction to Vector Search Algorithms k-NN
7:11 Implementing Vector Search in Python
11:08 Performance at Scale: k-NN vs. ANN Algorithms
17:15 Weaviate’s Custom HNSW Implementation
19:23 Multimodal and Multilingual Vector Search
22:15 Hybrid Search: Combining Keyword and Vector Search
24:01 Challenges of Scaling Vector Search
28:44 Demo: Implementing Vector Search with Weaviate
35:06 Streamlit Integration for Vector Search
41:26 Advanced Applications and AI Agents
44:09 Audience Q&A and Closing Remarks