[AI] Taulli T., Anderson B., de Vries J. - Building a Data and AI Platform with PostgreSQL [2026, PDF/EPUB, ENG]

Building a Data and AI Platform with PostgreSQL
Год издания: 2026
Автор: Taulli T., Anderson B., de Vries J.
Издательство: O’Reilly
ISBN: 979-8-341-62152-7
Язык: Английский
Формат: PDF (conv)/EPUB
Качество: Издательский макет или текст (eBook)
Интерактивное оглавление: Да
Количество страниц: 156
Описание: In a world where data sovereignty, scalability, and AI innovation are at the forefront of enterprise strategy, PostgreSQL is emerging as the key to unlocking transformative business value. This new guide serves as your beacon for navigating the convergence of AI, open source technologies, and intelligent data platforms. Authors Tom Taulli, Benjamin Anderson, and Jozef de Vries offer a strategic and practical approach to building AI and data platforms that balance innovation with governance, empowering organizations to take control of their data future.
PostgreSQL has some key technological advantages that naturally mix with the needs for GenAI development and platform creation:
• PostgreSQL has robust support for JSON and JSONB so it can handle rich data that sits at the heart of GenAI development, deployment, and long-term success. This is especially the case as models get more and more advanced and can infuse semi-structured data including embeddings, model parameter changes, or new configuration files. Imagine building a copilot or agent that you want to be more intelligent and capable of detailed personalization. Over time, they will continue differentiating for users and/or customers.
• The idea of extensibility is vital. Ideas such as semantic search are essential for effective GenAI design. It’s what drives personalization and the associated upsides in value for the data. This means users or developers can define their own functions, data types, and operators. It’s like having a dictionary with you that can cover any possible conversation in nearly any style or language.
• GenAI requires partitioning, indexing, and an efficient ability to query the large datasets critical for GenAI models. Without it, the learning curve and depth become too painful to deliver, and the associated tech debt increases. This is especially true with time-series data.
• The ability to seamlessly integrate with various programming languages (e.g., Python, Java, and C++) as well as libraries and heavily used AI frameworks (e.g., TensorFlow), also means you will have wider access to skilled labor than other environments. PostgreSQL can also serve as a backend for feature stores, providing storage for preprocessed data used by AI models. This is part of the living ecosystem of capabilities that are important for GenAI success.
• GenAI applications inherently rely on high data integrity and consistency. PostgreSQL’s ACID (atomicity, consistency, isolation, and durability) compliance ensures reliable transactions and data accuracy.
• PostgreSQL supports advanced SQL features, including common table expressions (CTEs), window functions, and full-text search, which are useful for data exploration and preprocessing in AI workflows.
• PostgreSQL has a vast and active open source community, contributing tools and extensions relevant to AI use cases. EDB is the leading company for investing in the complete health of PostgreSQL as a current and future-ready platform, with its contributions to the core codes and how it supports the ecosystem, nurtures PostgreSQL with components such as patching, and augments its capabilities for enterprise-grade performance. The community is constantly adding extensions such as PostGIS (for geospatial data) and HyperLogLog (for approximate distinct counting) to expand its utility for specific AI domains. No other open source database or other database does this.