AI is everywhere in developer conversations—but much of what’s discussed is either speculative or disconnected from real-world workflows. This talk offers a practical perspective from a working developer who uses AI tools daily—not as a researcher, but as someone shipping real software.

We’ll explore how AI is actually changing the development process in practice. Instead of writing every line of code manually, developers are increasingly guiding, reviewing, and refining AI-generated output. That shift sounds simple, but it has real implications for how you structure problems, validate solutions, and maintain quality.

You’ll see concrete examples of patterns that work: spec-driven development, breaking problems into AI-friendly chunks, and using tools effectively without becoming dependent on them. We’ll also cover where AI falls short—subtle bugs, overconfidence, and gaps in context—and how to catch those issues early.

This session includes a look at the actual tools and workflows I use, along with how my approach to problem-solving has changed over the past few years. The goal isn’t to convince you that AI replaces developers—it’s to show how to use it deliberately to move faster, stay focused, and produce better results without adding noise or complexity.