Integrating AI Tools into Your Development Workflow: A Practical Hands-on Workshop

AI-powered development tools have moved from novelty to necessity. Whether you're debugging existing code, scaffolding a new project, or navigating an unfamiliar codebase, your ability to leverage AI assistants effectively shapes both your productivity and the quality of your output. Yet for many developers, these tools remain underutilized, limited to autocomplete suggestions when they're capable of so much more.

Popular discourse often frames AI coding tools as enablers of "vibe coding," letting non-programmers generate applications through natural language alone. But this framing undersells their true utility. The full potential of these tools is unlocked not by those who avoid programming but by those who understand it deeply. Experienced developers who grasp algorithms, system design, and software engineering principles can leverage AI assistants to amplify their capabilities in ways that novices cannot: recognizing when generated code is subtly wrong, knowing which architectural patterns to request, and understanding how to decompose complex problems into AI-tractable pieces.

This course is designed for programmers who want to become better programmers. We assume familiarity with core programming fundamentals data structures, algorithmic complexity, version control, and software engineering practices. This is not a "build an app without coding" workshop. It is a systematic exploration of how AI tools can enhance the workflow of developers who already know how to code and want to work more effectively.

Through a combination of interactive instruction, live coding exercises, and real-world project work, you'll build skills that transform how you write, review, and ship code.

Understanding AI tools isn't the same as using them effectively. This workshop provides a structured laboratory for developing your skills. You will engage in progressive exercises designed to build competence and intuition, allowing you to:

- Navigate the landscape of AI development tools and make informed choices about which to adopt
- Build effective context, so AI assistants understand your codebase and constraints
- Develop workflows that leverage AI capabilities while maintaining code quality and security

Workshop Benefits
Participants will have the opportunity to develop practical skills to:
- Evaluate AI tools systematically by understanding the spectrum from autocomplete to agentic workflows, and selecting appropriate tools based on task requirements, privacy considerations, and organizational policy
- Construct effective context by learning techniques for helping AI assistants understand existing codebases, project conventions, and domain-specific requirements
- Design testable workflows that give AI clear success criteria through test-driven development practices, enabling confident iteration and verification
- Implement guardrails and automation using hooks for commits, linting, and quality checks that maintain standards without slowing development
- Scale AI-assisted development through parallel workflows, git worktrees, and multi-instance patterns that maximize throughput while maintaining coherence
- Extend capabilities strategically by configuring plugins, MCPs, and context management to tailor AI tools to your specific development environment
 

Topic Outline
Session 1: Foundations - Understanding the Landscape and Getting Started
- Taxonomy of AI development tools: copilots, chatbots, terminal assistants, agentic workflows
- Text-only vs. multimodal interaction models
- Agentic orchestrators as an emerging category
- Matching tools to tasks
- Hands-on: Claude Code (terminal and browser), Claude on claude.ai with Artifacts and code execution

Session 2: Building Context - Helping AI Understand Your Codebase
- Why context quality determines assistance quality
- Guiding AI through complex project structures
- Creating CLAUDE.md files and conventions documentation
- Building contextual scaffolding
- Hands-on: Context-building exercises with real codebases

Session 3: Testing and Tractability - The TDD Workflow
- Why TDD pairs naturally with AI assistance
- Tests as unambiguous success criteria
- The AI-adapted TDD loop
- Managing the iteration cycle between failures and modifications
- Building confidence through systematic verification

Session 4: Hooks and Skills - Automated Quality Assurance
- The power of linting & AI
- Custom skills for project-specific conventions
- Claude Code & Git Hooks
- Hands-on: Hooks, linter integration, custom skill configuration

Session 5: Multi-Clauding - Scaling AI-Assisted Development
- Parallel work environments
- Running multiple AI instances simultaneously
- Shifting from active coding to monitoring and review
- Task decomposition and work assignment strategies
- Merge strategies for parallel AI work
- Hands-on: Setting up & working with parallel development environments

Session 6: Plugins, MCPs, and Context Management - Extending Capabilities
- Claude Plugins Overview
- Model Context Protocol (MCP) Overview
- Considering MCP Context Overhead
- Evaluating when custom extensions are worthwhile
- Hands-on: Setting up a plugin or MCP