Over the past year, large language models have undergone a dramatic evolution. What once felt like “autocomplete on steroids” has matured into deeply capable AI engineering partners able not only to generate code, but to reason, verify, test, and collaborate across entire software lifecycles. Capabilities that seemed out of reach in early 2025 have quietly become everyday tools for AI-first engineers.

This talk is your fast-track guide to what changed, why it matters, and how to take advantage of it. We’ll explore the newest abilities of modern LLMs, how they enable more reliable workflows through automated verification and test-driven reasoning, and why practices like spec-driven development (SDD) are reshaping the way high-velocity teams build software. If you want to stay current with the latest trends in AI-assisted coding, and start applying them effectively, this session will bring you fully up to speed.