Getting Hands-On with Claude Code & Agentic AI Patterns

Agenda

• Getting Hands-On with Claude Code
• Agentic AI Patterns

Getting Hands-On with Claude Code

Claude Code is Anthropic’s developer-focused assistant for coding, refactoring, and working directly inside your projects. In this workshop, we’ll go beyond the hype and actually put it to work.

Who should attend and why?

• Senior & Lead Engineers: Learn how Claude Code can accelerate deep work—whether that’s refactoring legacy systems, exploring new codebases, or scaffolding greenfield features.
• Engineering Managers & Tech Leads: See how AI tools fit into your value stream. Understand what they really can and can’t do for your teams, and where the leverage points are.
• Founders & Product Builders: Cut through the noise on AI coding tools. Get a practical sense of how Claude Code helps you move faster without sacrificing quality.
• Craftspeople at heart: If you care about software craftsmanship, you’ll see how Claude Code supports (or challenges) practices like clean code, test-first, and value-driven design.

What we’ll do:

• Walk through real workflows with Claude Code (setup, commands, examples).
• Pair-program with it on live problems.
• Compare its “fit” in the broader toolchain (GitHub Copilot, Cursor, etc.).
• End with open Q&A and peer discussion—because the future of engineering is collaborative.

Why this matters:
AI is changing how we build software, but most conversations stay theoretical. This is a chance to ground it in practice, test it in a safe space, and share insights with peers who care about value as much as velocity.

Agentic AI Patterns

We’ll use the 17 agentic AI patterns as a mirror to look back at long-standing development principles. The idea isn’t to treat these patterns as brand-new magic, but to see how they map to fundamentals we already know:

• Separation of concerns → shows up in “multi-agent orchestration” and “role specialization.”
• Design patterns & modularity → echoed in reusable agent scaffolds like “retrieval augmentation” or “reflection/reflexive loops.”
• Feedback loops & testing → connected to patterns like “self-evaluation,” “sandboxing,” and “reward shaping.”
• Value streams & flow → mirrored in the choreography of agent collaboration, where bottlenecks and over-engineering look a lot like those in traditional pipelines.
• Resilience & fault tolerance → visible in ensemble/consensus patterns and redundancy.