Ensuring data + AI reliability through observability : aligning people, processes, and technology to drive value / Barr Moses and Michael Segner.

Format:

Book

Author/Creator:

Moses, Barr, author.

Segner, Michael, author.

Language:

English

Subjects (All):

Observers (Control theory).

Artificial intelligence.

Physical Description:

1 online resource (44 pages)

Edition:

[First edition].

Place of Publication:

Santa Rosa, CA : O'Reilly Media, Inc., [2025]

Summary:

So you're planning to invest in AI this year. Is your data AI-ready? Two-thirds of companies don't think so. As models become a commodity, the true competitive advantage when it comes to AI is reliable first-party data, and lots of it. This foundational report breaks down the key principles of data + AI reliability, with observability rather than traditional testing now the cornerstone of success. In Ensuring Data + AI Reliability Through Observability, you'll explore the common challenges involved in scaling retrieval-augmented generation models and learn how to overcome them with proven strategies. Authors Barr Moses and Michael Segner show you how reliability impacts performance, and they provide practical procedures for making better-informed decisions about your team's time and resources. Leverage structured and unstructured data to exponentially increase the value and complexity of AI applications Understand the principles behind building reliable systems and how they apply to data and AI products Accelerate and scale reliable data and AI products initiatives Overcome common reliability challenges with building data and AI products at scale Prioritize your team's time and resources to maximize data and AI reliability.

Notes:

OCLC-licensed vendor bibliographic record.

ISBN:

9798341649453

OCLC:

1540344968