Explore binary quantization methods for vector databases in this technical talk featuring Qdrant CTO Andrey Vasnetsov. Dive deep into optimization techniques and search speed improvements for vector stores, addressing the challenges posed by expensive vector searches and large neural network encoders. Learn how binary quantization serves as a solution for vector compression, particularly when dealing with low-latency storage requirements and scalability concerns. Master practical implementations of vector indexing, understand the benefits of oversampling, and discover the compatibility of various models with binary quantization. Compare dot product and Hamming distance calculations while gaining insights into real-world applications from an experienced Machine Learning Engineer who emphasizes practical demonstrations over theoretical concepts.