Claude at Work

The Claude suite of LLM-powered tools and agents from Anthropic has radiacally–and permanently–chamnged the way technical professionals design and create software. From agentic code generation to assistants that capture and execute specifications, assist with architectural and visual design, automate testing, and much more, Claude is a hyper-productive working partner. Claude at Work teaches you how to seamlessly step into your new role as AI agent manager, turning Claude’s tools and models into a full DevOps team and ensuring you stay in full control of the process.

In this practical book written for busy software engineers with responsibilities along the full SDLC, you'll build a collaborative workspace application—similar to a hybrid of Slack and Figma—that expands in scope as you progress. As you explore this collaborative AI workflow, you’ll discover how to leverage Claude Cowork to securely access and edit files directly in your project directories. When you scale the system, you’ll see how to direct Claude Code on parallel workstreams, focusing your time on task design, validation, and architectural approvals.

As you progress, you’ll put Claude to work designing architecture and components. Everything you’ll do is grounded in established software engineering fundamentals, including behavior-driven development, test-driven development, and rigorous documentation, so you can seamlessly connect new AI techniques to your existing team workflows. You’ll see how to write reusable skills to automate repetitive administrative tasks, how to maintain a healthy skepticism toward AI-generated output, and how to manage the real-world token costs of running agent sessions.

You’ll even find practical exercises that empower you to build custom AI agents that serve as an autonomous operations team. These specialized agents run alongside your application in production to deploy, monitor, and repair live systems, while you learn exactly how the platform’s internal safeguards work to prevent rogue activity.