Mohlo by se vám také líbit

Více informací o knize Production-Ready Multimodal Data Engineering

• Parametry knihy
• Anotace
• Oblíbené z jiného soudku

Nákupem získáte 43 bodů

Anotace knihy

AI applications are only as reliable as the data infrastructure behind them. Are you ready to engineer systems that go beyond prototypes?

Building enterprise AI platforms requires more than Python scripts, vector databases, and language models. You need reliable enterprise AI data pipelines, scalable processing, consistent storage, secure retrieval, and the operational controls that keep complex systems working together.

If you're struggling with unstructured data processing using Python, designing real-time streaming data architectures, or connecting Apache Spark, Kafka, lakehouse technologies, and multimodal RAG into one coherent system, this book provides the practical engineering direction you need.

Production-Ready Multimodal Data Engineering guides you through the architecture and implementation of an Enterprise Multimodal Data Platform (EMDP). Through practical Python examples and progressively integrated workflows, you'll learn how to build the foundations of scalable AI data infrastructure while understanding the engineering decisions required for reliable enterprise systems.

Inside, you'll learn how to:

• Engineer enterprise AI data pipelines for documents, images, audio, video, and structured information using Python, Apache Kafka, and reliable ingestion techniques.

• Master unstructured data processing with Python, including document extraction, image processing, audio transcription, video frame extraction, metadata validation, and distributed transformations.

• Build modern data lakehouse architectures using Apache Spark, Delta Lake, and MinIO to organize, process, and manage multimodal datasets.

• Design real-time streaming data architectures with Apache Kafka, event-driven processing, checkpointing, deduplication, and failure recovery strategies.

• Implement vector search, indexing, and retrieval using Qdrant, text and image embeddings, similarity search, and cross-modal retrieval techniques.

• Develop enterprise multimodal RAG applications with hybrid semantic search, reranking, grounded responses, and FastAPI-based knowledge retrieval.

• Automate complex data workflows through Apache Airflow workflow orchestration, scheduled processing, backfills, and operational recovery.

• Strengthen security, governance, and observability with authentication, authorization, data quality controls, lineage tracking, automated testing, metrics, and distributed tracing.

• Explore production deployment and MLOps practices involving Docker, CI/CD, model versioning, performance optimization, cloud scalability, and disaster recovery planning.

More than a collection of disconnected tutorials, this book provides a practical engineering blueprint for connecting modern data technologies into a unified multimodal AI platform.

Whether you're a data engineer, machine learning engineer, AI developer, or software architect, you'll develop the technical understanding needed to design dependable data pipelines, manage multimodal information, and build enterprise AI applications supported by reliable retrieval infrastructure.

Move beyond isolated AI experiments. Master the engineering principles behind scalable multimodal data platforms and build the foundation for production-oriented enterprise AI systems.

Get your copy today and start engineering the infrastructure behind modern AI.

Parametry knihy

• Plný název: Production-Ready Multimodal Data Engineering
• Podnázev: Build Enterprise AI Data Platforms with Python, Spark, Kafka, Lakehouse Architecture, Vector Databases, Multimodal RAG, and MLOps
• Autor: Lucan Verid
• Jazyk: Angličtina
• Vazba: Brožovaná
• Počet stran: 144
• EAN: 9798140623647
• ID: 54439234
• Nakladatelství: Independently published
• Hmotnost: 352 g
• Rozměry: 280 × 216 × 8 mm
• Datum vydání: 07. November 2026

---

---

---