Hello!
Today I'm thinking about job hunting, specifically that the world you're applying for a job in no longer exists. There are remnants of it in bigger, older companies, but not for long. Why? Because the job descriptions most people are applying against assume a workflow that's already gone. For years, designers worked "out ahead" of engineering in mockups. Now development teams are "out ahead" of design, with new features becoming code at a furious pace, and design's job shifts toward making all that output fit together. At the Design Futures Assembly this spring, close to half of designers had shipped AI-generated code to production, and at the best teams the "stack" was dozens of internal tools that change month to month. Try writing a stable job req for that. So the natural reaction is to grab every tool and try to do every role. Designers pick up coding agents, developers spin up layouts, and everyone ends up here: That approach spreads us thinly across multiple complex and important jobs. I can play a developer on TV, but in real life I'm a pretty crappy one. The better move is the lesson we taught designers at Yahoo! years ago: double down on the thing you're uniquely good at and apply it elsewhere. So if you're looking for work right now, don't optimize for the job posting. Get hands-on with the tools, figure out what they can and can't do, and show how your particular superpower reaches further because of them. That's the job that actually exists.
How can AI change the way we design onboarding for empty software states?
Anyone who's designed software has hit the "empty state" problem. Spreadsheets, presentation tools, document editors, design tools, project management apps... they all face the same hurdle: how do you help people get started when they're staring at a blank canvas?
Our usual answers are coach marks, templates, tours, overlays, and videos. The problem? These approaches require people to learn first, then act. But in reality most of us just jump right into doing and only fall back on learning if what we try doesn't work. Shockingly, asking people to read the manual first doesn't work.
So what's different now? AI can model how to use a product by actually going through the process of creating something and letting people watch. From there, people can just tweak the result to get closer to what they want. AI does the creation, people do the curation. Rather than learning then doing, people observe then refine.
We built this into ChatDB. When you drag a spreadsheet onto the page, the system picks a color, picks an icon, names the dashboard, starts running analysis, and generates charts for you. It even pins a chart to the dashboard, which quietly shows you that pinning exists. You can change or undo all of it, but by then you've already seen the capabilities in action.
The part I like most is how the AI's choices do double duty: they're useful defaults and implicit feature discovery at the same time. Once your dashboard is made, it's trivial to edit the title (just click and type), change the icon, colors, and more. AI gives you the starting point and you take it from there.
Effectively, the traditional empty state problem transforms from "how do we help people start?" to "how do we help people refine?" So next time you reach for a tooltip tour, ask whether the product could just do the first piece of work itself and let people take it from there.