AI.js Day 2026

Every prompt branches into a hundred possible outcomes – and the demo only ever shows you the one that worked. Danny digs into what's lurking in the branches you don't see: the diffs nobody fully reads, the edge cases that only surface in production, the gap between "it compiled" and "it's actually right".

Expect a practical look at verification debt, review habits, and the engineering discipline it takes to make sure the branch that ships is the one you actually meant to build.

If AI keeps taking over more of the implementation work, which areas remain the most valuable for human contributions? In this talk, we'll take one step back from the hypothetical end and focus on the durable skill that has always separated great engineers from merely productive ones: judgment.

We'll talk about product engineering, accountability, trade-offs, constraints, evaluation, and how to keep making software worth having in an AI era.

How do you even become an AI engineer? We’ll go through what the job looks like day to day, and the practical path to get there from a frontend/JS background: how much programming, math, and ML you really need, and how to learn more complex concepts effectively. We’ll also touch on how to develop a sustainable practice to continual learning so you stay competitive for the long haul.

Design sources, components, browser, IDE – many tools, one developer trying to get software out the door. Figma, Chrome DevTools, Storybook, MUI, Callstack, and WebStorm raise a harder question to ponder: not whether AI helps or hurts, but how to actually engineer these pieces together into a workflow that scales and remains worth using. A live look at where over-automation is quietly costing you control and where it’s genuinely adding value.