Название: Advanced SQL: Implementing Modern Data Solutions and ML Applications (Final Release)
Автор: Rui Machado, Hélder Russa, Pedro Esmeriz
Издательство: O’Reilly Media, Inc.
Год: 2026
Страниц: 482
Язык: английский
Формат: pdf, epub
Размер: 10.1 MB

SQL has evolved from a querying language for relational databases to a foundational tool for building scalable, modern data solutions across real-time analytics, machine learning workflows, and even generative AI applications. Advanced SQL moves beyond SELECT statements and taps into the full power of SQL as a programming interface for today's most advanced data platforms.

Written by seasoned data experts Rui Machado, Hélder Russa, and Pedro Esmeriz, this practical guide explores the role of SQL in streaming architectures (like Apache Kafka and Flink), data lake ecosystems, cloud data warehouses, and ML pipelines. Data and analytics engineers, analysts, and scientists will get hands-on guidance on expanding their SQL skills into emerging workflows and real-world production systems.

Use SQL to design and deploy modern, end-to-end data architectures
Integrate SQL with data lakes, stream processing, and cloud platforms
Apply SQL in feature engineering and ML model deployment
Master pipe syntax and other advanced query features
Leverage SQL to build GenAI-ready data applications and pipelines

This book covers advanced SQL not as a historical artifact but as the living language of modern data systems, the substrate on which analytical platforms, real-time pipelines, and machine learning applications are built. When you understand how window functions track state across ordered partitions, you will know when the generated code handles ties incorrectly. When you understand the mechanics of incremental processing, you will catch the moment an AI-suggested approach duplicates records under late data. When you understand how SQL integrates with ML workflows, feature stores, training pipelines, and model scoring, you will identify when a well-intentioned query has introduced a leak between training and serving that will surface only when the model is in production.

This book does not aim to make you faster at writing SQL. Rather, its goal is to make you the kind of practitioner who can be trusted: with complex systems, with data that drives consequential decisions, with the ML applications that increasingly shape what organizations know about the world. In a moment when code can be written almost on demand, that trust is the rarest and the most valuable thing you can offer. Build systems worth trusting, guide the machines well. That has always been the work, and it still is.