- ホーム
- > 洋書
- > 英文書
- > Computer / General
Full Description
Build a user-facing AI product from a validated problem to production, combining RAG, agents, deployment, user acquisition, and iteration in one cohesive project
Key Features
Build one cohesive RAG and agent product from a validated idea to production
Use principles-first guidance to make sound choices across fast-changing AI tools
Go beyond the prototype with deployment, user acquisition, evaluation, and iteration
Book DescriptionMany AI projects work as demos but never become products that people use. This book focuses on the full product lifecycle: choosing a worthwhile problem, defining a scoped solution, building the application, deploying it, finding users, and improving it with real feedback.
You begin by validating demand, mapping the user journey, and selecting an architecture that fits the problem. From there, you build a RAG-based assistant in Python, add tool use and function calling for agent workflows, and make informed choices across LLMs, vector stores, and frontend options. The project then moves into production with cloud infrastructure, CI/CD, access controls, and cost guardrails. The final chapters address acquisition, pricing, structured logging, LLM evaluation, prompt versioning, user behavior, and post-launch paths.
By the end, you will understand how the RAG and agent stack fits into a complete AI product and how to move from an idea to a deployed, user-facing application you can continue to improve.What you will learn
Validate an AI product problem before committing to a build
Design a scoped product journey and choose the right architecture
Build a RAG assistant with Python and an LLM API
Add tool use and function calling to create agent workflows
Deploy an AI application with CI/CD and cost controls
Acquire early users and gather structured product feedback
Evaluate LLM outputs and manage prompts in production
Prioritize product improvements using real user behavior
Who this book is forThis book is for product managers, AI strategists, engineers, analysts, and developers who want to move beyond AI prototypes and understand the complete product lifecycle. Basic Python programming is expected; prior AI/ML, frontend, or cloud infrastructure experience is not required.
Contents
Table of Contents
Why Most AI Projects Never Become Products
Finding a Problem Worth Solving
Designing the Product Before You Code
Choosing Your AI Building Blocks
Building the MVP - From API Call to Working App
Adding the Agent Layer
Shipping It - Production Infrastructure That Actually Works
Getting Your First 10 Users
Iterating With Real Users
From Side Project to Real Product
-
- 電子書籍
- 剣聖令嬢はヤンデレ賢者に執着される ~…
-
- 電子書籍
- 気持ちイイ恋 教えて下さい~下手なあな…
-
- 電子書籍
- 未知なるリンパ - 全身にはり巡らされ…
-
- 電子書籍
- 看取り、できますか? - 患者・家族が…



