LLM Code Workflows

The rise of large language models (LLMs) like Claude, ChatGPT, and specialized tools like Windsurf has fundamentally changed how developers work. This conic explores practical patterns for integrating AI into your daily coding practice—not as a replacement for thinking, but as a powerful amplifier of productivity and creativity.

What You'll Find Here

Pair Programming with AI

• How to structure prompts for maximum code quality
• When to trust AI suggestions and when to verify
• Debugging strategies using LLM assistance
• Building mental models of AI capabilities and limitations

Code Generation Patterns

• Scaffolding boilerplate with precision
• Generating test suites and documentation
• Refactoring and optimization workflows
• Type-safe code generation in TypeScript

Productivity & Workflow

• Integrating AI tools into your IDE (Windsurf, Copilot, etc.)
• Building custom prompts and workflows
• Maintaining code quality while moving faster
• Balancing automation with intentionality

Real-World Case Studies

• How I use Claude for architecture decisions
• Windsurf workflows for Next.js projects
• LLM-assisted accessibility audits
• Building this portfolio with AI assistance

Why This Matters

LLMs are not a fad—they're a fundamental shift in how knowledge work happens. As a developer, you have two choices:

1. Learn to work with them effectively and multiply your impact
2. Ignore them and watch your productivity plateau

This conic is about the first path. We'll explore the practical, the philosophical, and the tactical—because using LLMs well requires all three.

---

Browse the articles below to dive deeper into specific workflows and techniques.