Isbn: 9798194216451 - qdrant: the complete guide to vector search and semantic ai applications: store embeddings, build hybrid search, filter with payloads, and deploy collections on cloud and self-hosted (4 risultati)

PAP. Condizione: New. New Book. Shipped from UK. Established seller since 2000.

Taschenbuch. Condizione: Neu. Neuware - Build accurate, scalable semantic search systems that go far beyond basic nearest neighbor retrieval.Modern AI applications need more than a place to store embeddings. They need hybrid retrieval, metadata filtering, reranking, access control, efficient indexing, measurable relevance, reliable ingestion, and production infrastructure that can scale with real workloads.This practical guide takes you from vector search fundamentals through sophisticated retrieval pipelines and production operations, showing how Qdrant fits into RAG, enterprise search, recommendation systems, agent memory, multimodal applications, and other semantic AI workloads.What you will learn: - Design collections with dense sparse named and multivector representations- Generate and store embeddings with FastEmbed and multiple inference workflows- Build precise payload filters nested conditions indexes facets and access controls- Use the unified Query API for similarity search recommendations discovery grouping sampling and multi stage retrieval- Implement native BM25 sparse retrieval full text search and tenant aware lexical relevance- Combine dense and sparse candidates with RRF weighted RRF and distribution based score fusion- Build reranking pipelines with ColBERT cross encoders MUVERA and Matryoshka representations- Apply Formula Queries for freshness popularity geography and business aware ranking- Measure Recall Precision MRR NDCG ANN accuracy latency and retrieval quality- Build production RAG with chunking source grouping metadata and permission filters- Create product search enterprise search recommendations agent memory multimodal search and code search systems- Tune HNSW filtered ANN ACORN memory tiers and optimizer behavior for demanding workloads- Reduce memory and storage costs with scalar binary product TurboQuant and Turbo4 compression- Manage bulk ingestion retries incremental embedding updates model migrations and data lifecycles- Scale with shards replicas consistency controls custom sharding tiered multitenancy and capacity planning- Deploy and operate secure systems with Docker Kubernetes Helm managed cloud hybrid cloud private cloud monitoring backups and disaster recoveryYou will also see how advanced retrieval components fit together, including candidate prefetch, global fusion across shards, late interaction reranking, business aware scoring, tenant isolation, read write contention, quantized rescoring, snapshots, migrations, and embedded edge retrieval.Hands on Python code, API examples, Shell commands, and configuration samples connect the concepts to implementation so you can apply them to real semantic search and AI retrieval projects.Grab your copy today and build retrieval systems designed for relevance, scale, and production use.…

Condizione: New. Print on Demand.

Paperback. Condizione: new. Paperback. Build accurate, scalable semantic search systems that go far beyond basic nearest neighbor retrieval.Modern AI applications need more than a place to store embeddings. They need hybrid retrieval, metadata filtering, reranking, access control, efficient indexing, measurable relevance, reliable ingestion, and production infrastructure that can scale with real workloads.This practical guide takes you from vector search fundamentals through sophisticated retrieval pipelines and production operations, showing how Qdrant fits into RAG, enterprise search, recommendation systems, agent memory, multimodal applications, and other semantic AI workloads.What you will learn: Design collections with dense sparse named and multivector representationsGenerate and store embeddings with FastEmbed and multiple inference workflowsBuild precise payload filters nested conditions indexes facets and access controlsUse the unified Query API for similarity search recommendations discovery grouping sampling and multi stage retrievalImplement native BM25 sparse retrieval full text search and tenant aware lexical relevanceCombine dense and sparse candidates with RRF weighted RRF and distribution based score fusionBuild reranking pipelines with ColBERT cross encoders MUVERA and Matryoshka representationsApply Formula Queries for freshness popularity geography and business aware rankingMeasure Recall Precision MRR NDCG ANN accuracy latency and retrieval qualityBuild production RAG with chunking source grouping metadata and permission filtersCreate product search enterprise search recommendations agent memory multimodal search and code search systemsTune HNSW filtered ANN ACORN memory tiers and optimizer behavior for demanding workloadsReduce memory and storage costs with scalar binary product TurboQuant and Turbo4 compressionManage bulk ingestion retries incremental embedding updates model migrations and data lifecyclesScale with shards replicas consistency controls custom sharding tiered multitenancy and capacity planningDeploy and operate secure systems with Docker Kubernetes Helm managed cloud hybrid cloud private cloud monitoring backups and disaster recoveryYou will also see how advanced retrieval components fit together, including candidate prefetch, global fusion across shards, late interaction reranking, business aware scoring, tenant isolation, read write contention, quantized rescoring, snapshots, migrations, and embedded edge retrieval.Hands on Python code, API examples, Shell commands, and configuration samples connect the concepts to implementation so you can apply them to real semantic search and AI retrieval projects.Grab your copy today and build retrieval systems designed for relevance, scale, and production use. This item is printed on demand. Shipping may be from our UK warehouse or from our Australian or US warehouses, depending on stock availability.…