Architected Intelligence: Principles for Building AI-First Organizations and Technologies
暫譯: 架構智能：建立以AI為首的組織與技術原則
Miller, Jacob, Mumford, Jeremy
相關主題
商品描述
Transform AI potential into organizational results through the Applied Intelligence framework
More than 80% of AI projects fail - twice the rate of traditional IT initiatives. Architected Intelligence: Principles for Building AI-First Organizations and Technologies bridges the gap between executive vision and practical implementation by providing technical leaders with timeless frameworks that remain relevant as technology evolves. Drawing on their experience leading Pattern to become the world's largest ecommerce accelerator, authors Jacob Miller and Jeremy Mumford deliver actionable guidance for turning proofs of concept into production systems.
Architected Intelligence organizes AI success around five core components:
- Design AI Systems to Deliver Impactful Output
- Power AI with High-Quality Input Data
- Engineer, Optimize, and Integrate AI Models
- Create Trust through Robust Observability
- Scale Transformation through Effective AI Enablement
Readers will discover the Service as a Software mindset to build AI-first capabilities, best practices to consistently manage context engineering at scale, and a clear playbook to maximize user trust by disaggregating trust into its underlying elements so each can be intentionally strengthened.
The book also provides:
- Practical guidance on building AI workflows and agents incrementally, avoiding the all-or-nothing trap that cause project failures
- A democratization of expertise framework, including comprehensive unstructured data governance strategies including knowledge acquisition, curation, and AI-optimized search methods
- Repeatable processes for AI provider evaluation, model selection, hosting decisions, and advanced configurations
- Implementation strategies for monitoring inputs and outputs, establishing security and privacy controls, tracking resources, and building accessible AI foundations
- Guidance on the foundational platforms necessary to scale AI systems across an entire organization
Perfect for technical CEOs, CTOs, leaders in Data Science and AI, Directors of Engineering, and others shaping or executing AI strategies, Architected Intelligence serves as a foundational reference for decision makers who want to establish best practices and inspire the transformations necessary to sustainably harness AI's potential.
商品描述(中文翻譯)
透過應用智慧框架將人工智慧潛力轉化為組織成果
超過80%的人工智慧專案失敗，這是傳統IT計畫的兩倍。架構化智慧：建立以AI為先的組織和技術的原則彌補了高層願景與實際實施之間的差距，為技術領導者提供了隨著技術演變而持續相關的永恆框架。作者Jacob Miller和Jeremy Mumford利用他們將Pattern打造成全球最大的電子商務加速器的經驗，提供可行的指導，將概念驗證轉化為生產系統。
架構化智慧圍繞五個核心組件組織人工智慧的成功：
- 設計AI系統以提供有影響力的輸出
- 以高品質的輸入數據為AI提供動力
- 工程化、優化和整合AI模型
- 通過穩健的可觀察性建立信任
- 通過有效的AI啟用擴大轉型
讀者將發現服務即軟體的思維模式，以建立以AI為先的能力，最佳實踐以一致地管理大規模的上下文工程，以及一個清晰的行動計畫，通過將信任分解為其基本元素來最大化用戶信任，以便每個元素都能被有意識地加強。
本書還提供：
- 關於逐步建立AI工作流程和代理的實用指導，避免導致專案失敗的全有或全無陷阱
- 專業知識民主化框架，包括全面的非結構化數據治理策略，包括知識獲取、策展和AI優化搜索方法
- 可重複的AI供應商評估、模型選擇、託管決策和高級配置的流程
- 監控輸入和輸出、建立安全和隱私控制、追蹤資源以及建立可訪問的AI基礎的實施策略
- 關於在整個組織中擴展AI系統所需的基礎平台的指導
本書非常適合技術CEO、CTO、數據科學和AI領域的領導者、工程總監以及其他塑造或執行AI策略的人士，架構化智慧作為希望建立最佳實踐並激發必要轉型以可持續利用AI潛力的決策者的基礎參考。
作者簡介
JACOB MILLER is the Vice President of Data Science at Pattern, the world's largest e-commerce accelerator. He has assisted organizations and leaders implementing AI solutions with demonstrable and immediate benefits.
JEREMY MUMFORD is the Lead AI Engineer at Pattern. He brings together data and software engineering with formal training in databases and data science to build production-ready generative AI systems.
作者簡介(中文翻譯)
雅各·米勒是Pattern的數據科學副總裁，Pattern是全球最大的電子商務加速器。他協助組織和領導者實施具有可證明和立即效益的人工智慧解決方案。
傑瑞米·曼福德是Pattern的首席人工智慧工程師。他將數據和軟體工程結合，並擁有數據庫和數據科學的正式訓練，以構建可投入生產的生成式人工智慧系統。
