I just got back from 2026’s Figma Config, held in San Francisco, and I’m still sorting through everything I saw and heard while I was there. But after some rest and getting over the annual Config Cold 🤧, I’m excited to share many of my thoughts about what’s new from Figma with you, and a bit about where our industry feels like it’s headed right now.

Every year I go, the part that I look forward to most isn’t the keynote or the new features, as exciting as they may be. It’s the moments of running into someone I’ve only ever talked to online, catching up with a past student or friend I haven’t seen since last year, or meeting a designer who’s just landed their first design systems role and can’t wait to share more about it. I always leave feeling so grateful for this community of ours, and this year was no different.

I’ll be honest, though. The energy felt a little different this year. There was still excitement in the room (the audience just about exploded when multiple export options for animations were shown), but the truth is that a lot of us in design are a little worn out right now. Layoffs have touched nearly everyone I’ve talked to, whether it was them, their team, or a friend still looking for their next role. There’s a very real uncertainty hanging over our industry at the moment, and I have to admit it would feel strange to write a cheerful recap of everything new without first acknowledging the rooms those features are landing in. So before we dive in, I just want to say, if you’re feeling that weight too, you’re not alone in it. You have about 10,000 designers sitting right next to you.

So, with all of that out in the open, let’s get into what’s actually new. ✨

Figma framed this year around a single idea: the canvas as the place where everything comes together. Materials, such as motion, shaders (woo!), and more, all living in the same file as your components, your variables, and your team. It’s a big, but natural swing, and it feels like Figma is stepping further onto Adobe’s turf than ever. Whether every piece lands, I’m not sure, but I think that framing is worth holding onto as we go through it.

Code on the Canvas

For years, we’ve all heard some version of the “design versus code” debate, and even if you’re not quite sure where you landed on that one, one thing is probably clear: you’re tired of hearing about it. Most of our tools have either subtly or very intentionally forced us to pick a side. Figma’s take this year is that this was always a bit of a false choice. Code is just another material, just like an image or a vector, and it should be able to live right on the canvas like everything else.

So, that’s exactly what they’re introducing. Soon in Figma, we’ll be able to turn any design layer into an interactive code layer with a single click, or with a prompt, and then duplicate it to explore multiple directions side by side, the same way we’d do with a frame. From their demos, it feels very much like viewing and playing with a prototype, but instead of in a new window or tab, it’s right there on the canvas. Once that code layer exists on the canvas, you can work with it just as you would any other layer. Commenting, duplicating, annotating, and iterating with others who may be in the same file. When you want to go back the other way, you can pull a code layer back into editable design layers, make your changes, and then push it back with a click.

The part that really caught my attention here is being able to explore real directions side by side. That’s the good stuff. More on why in just a bit!

I haven’t been able to get my hands on this one just yet, as it’s rolling out in a closed beta next month. But be sure to join the waitlist if you haven’t already! The part that really does interest me, especially as a systems person, is less in the code itself and more in what happens to it. For a long time, code has lived in a very separate, single-player space, and it was one that, if I’m being honest, always felt very intimidating to play in. But pulling that code into the canvas, where the rest of us designers already work together, feels like the very meaningful and big change here.

Figma Motion

Here’s an honest confession: I’m not a very good motion designer. Heck, I’m terrible. I’ve always had such a hard time seeing in motion, if that makes sense. I can hold a static layout in my head all day, but ask me to imagine how it should move, and I freeze up. So, I’ll admit that I walked into this announcement a little unsure of how much it was for someone like me, or how useful it could be.

What Figma showed is a real timeline living right inside of Figma Design, with keyframes, presets, and the rest. You can build motion from scratch, layer it onto something you’ve already made, or even ask the Figma Agent for a starting point. And those export options that nearly brought the entire audience to its feet? They live here, too.

But the part that I keep thinking about is maybe something my fellow design systems friends will appreciate, too. Because motion is now built into the Figma platform, it can (and probably should) become part of your design system. We can now animate a component once, and that motion travels across every screen and in every teammate’s file, the same way our color fills and typography definitions already do. A motion primitive, basically. And because Figma now supports motion variables, I think it’s an area we’re all going to have to begin thinking about sooner than later.

Honestly, though, what I’m most excited about is smaller and more personal. For someone who’s always felt locked out of motion by my own brain, having it right there on the canvas feels like a very open invitation to finally play around with it more and more. No new tool to learn, no separate file to manage, basically, no excuse not to try. And since it’s already in open beta, I don’t even have to wait. Like most things, I have a feeling that the more I let myself play and explore with it, the less intimidating it’ll start to feel. Hey, maybe I’ll surprise myself a little with what I may be able to create.

Shaders

Even if the word is new, you’ve likely seen shaders used all across the web. They’re those grainy gradients, the liquid-looking surfaces, and the bits of light and texture that make something feel almost physical. They’ve actually powered Figma’s canvas rendering for years, but making your own meant writing in a language that most of us (besides Rogie) never learned. I very much filed them under “things other, way smarter people do.”

But now, you can describe the effect you want, or even hand Figma a reference image, and the Agent builds the shader for you. And because it’s built that way, what comes back isn’t locked in a black box. It arrives with controls right there on the canvas, which means you can adjust it, stack it with other effects, and shape it until it feels like yours.

And for as visually impressive as they are, what really stood out was how performant they felt on the canvas. The demos all ran smoothly and in real time. And of course, because this is Figma, it’s all happening on the web, available in both the desktop app and the browser.

The moment it really clicked for me was seeing shaders meet Motion. Every property a shader exposes can now be keyframed on the Motion timeline, which means anything you can nudge with a slider, you can also animate over time. In the demos, you could watch a shader ripple and shift across a design that was already moving, and even adjust it while the animation played. Seeing all of that happen on the canvas, in a single file, was the kind of thing that used to take a stack of specialized tools and a whole lot of patience.

And that’s the thread I keep pulling on. Between code, motion, and shaders, Figma is pouring real energy into one idea: raising the ceiling of what’s even possible to attempt on the canvas. They’re building us a taller room. What we actually make in it is still up to us, and I’ll have more to say on that in a bit.

Generative Plugins

Every designer I know has a handful of small, oddly specific tasks they do over and over in Figma. Renaming a batch of layers a certain way, laying out a set of images just right, or generating placeholder data based on the company they work for so components and designs appear more realistic. For years, the fix was always hoping that someone in the community had the same need, and built a plugin for that workflow.

And so often, someone had. I think this is worth pausing on, too. So much of what’s made Figma feel like Figma came from the community of designers and developers who built plugins and files in their spare time and gave them away, then quietly maintained them for years as the tool shifted underneath them. That’s a very real, and often thankless amount of work, and a lot of our daily workflows have been quietly held up by it. Before anything else here, I just want to say thank you to those people.

The new thing is you can now describe the tool you need, the behavior, the controls, the parameters, and build it without setting up a dev environment or any plugin API to learn. You can keep it for yourself, share it in your file, and soon publish it to your team, organization, or the wider community. As a systems person, I find that super interesting. We’re used to sharing components and patterns, but now we may start sharing our own internal tools, too.

I’ll admit I have an open question, though, and I don’t think I’m alone in it. If anyone can spin up a tool on demand, where does that leave the community plugins we’ve all relied on, and the people who made them? Maybe the answer is that the small one-off tools get easier to make yourself, while the deep, well-loved ones matter more than ever. I don’t know yet. But I hope, whatever happens, we don’t lose sight of how much those builders gave us along the way.

Figma Weave

This is the one I find most interesting from a strategy angle, especially now that Figma is a public company. Weave started life as Weavy, a small Tel Aviv startup barely a year old, which Figma acquired last fall in what was reportedly its largest deal ever, somewhere north of $200 million. For a roughly twenty-person team that had raised only a few million, that’s a big number, and it tells you how much Figma wanted what they’d built.

So, what did all of that money get them? Weavy is now Figma Weave, and it’s a node-based canvas for generative work. Instead of firing off a single prompt and hoping for the best, you connect models together, transform assets, compare different directions, and refine the output step by step in a very visual way. The idea is that any AI result isn’t the finish line, it’s material (there’s that word again) you shape. At Config, those tools started making their way directly onto the Figma canvas, too. It all comes back to the canvas this year.

I don’t have a tidy verdict on this one yet, and I think that’s okay. But I’ll say it lands squarely on the theme running through everything this year: exploration. A tool built entirely around trying things, branching, and comparing is about as on-message as it gets for “the canvas is where everything connects.” I’m really curious to see where it goes and how people end up using it.

The Figma Agent

If you’ve been keeping track, in almost every announcement and feature, the Figma Agent has shown up in some way. It’s what turns your description into a shader, generates a starting point for motion, and creates the plugin that’s just for your specific workflow. So it makes sense to end here, because in a lot of ways, Figma’s agent is now feeling like the thread that’s running underneath it all. It launched to everyone the day before Config, and at the show, it picked up a few new abilities that I think are worth knowing about.

The first is skills. You can finally package up your own workflows and conventions into reusable instructions that the Figma Agent follows, and then use ones your team has made or pull directly from the Figma Community. If you’ve ever written documentation hoping that people would follow it, this idea will feel very familiar, except now the guidance is something the agent can act on directly. The second big thing is connectors, which let the agent reach into the tools you already use, like Notion, Slack, GitHub, Granola, and others, and let that context flow into the canvas. Kind of cool, and I’m excited to try this!

Staying Wide

A quick thing before I wrap this post up, because I’ve promised a few times that I’d come back to it.

If there’s a single thread running through everything this year, I think it’s exploration. Code layers you can duplicate to try a few directions at once. Motion and shaders you can poke at without leaving for another tool. Plugins you can spin up for whatever task is in front of you. The entire pitch is that the canvas is, and will continue to be, a place to try things.

And I think that matters more than it may seem, because of something I keep noticing about AI. It’s genuinely great at helping us as designers move fast and efficiently, but it tends to pull us in deep before we’ve gone wide, and I think that’s a mistake. It’s become easier than ever to describe what you want, and to receive a polished version, often before we’ve really had a chance to sit with the problem. The tricky part is that the cost of this is invisible. We never see the ideas we didn’t explore, and we may not see the solution which might have been the better one for the problem we’re up against.

That’s the part I don’t want to lose. The messy middle, where we try the obvious thing, then a few worse ones, then the strange one. It’s the place where the best ideas are hiding. It’s also, for many of us, the most fun part of the job.

So here’s the quiet hope I’m carrying out of this year’s Config: use AI to go fast, and use the canvas to stay wide. Not one or the other, but both. I think the ceiling for what we can make gets raised by that messy middle, and by the people who are willing to stay in it to explore.

Which all brings me back to where I started. A lot of us are tired right now, and unsure of what the next year holds. I won’t pretend a new set of shiny and impressive features changes any of that, but if there’s something steadying in here, maybe it’s this: the tools will keep changing, the way they always have. What lasts is how we think, and the people we get to figure it out alongside. To those 10,000 designers sitting next to one another in San Francisco last week, and to you reading this, I think we’re going to be just fine. 💛

Our First Dinner

I can't wrap up without mentioning one more moment from the week, and it might be the one I'll remember longest. We hosted our very first Baseline Design dinner.

Getting to spend an evening with so many people we've admired over the years, all in one room, was really, really special. It still feels a little surreal that Baseline has grown into something that can bring folks like that together. I'm so grateful that everyone who joined wanted to spend their evening with us.

I'm really proud of our team, and honestly couldn't be more excited for what's ahead. ❤️

Growing the Bench

We’re very excited to be growing our contractor bench at Baseline Design!

We work alongside brand and product teams on the architecture, governance, adoption, and all the messy, human parts of making a design system successful. We care just as much about how teams work together as we do the components themselves.

If you’re excited about where design systems are going, whether you’re building in Figma, creating with AI, writing tokens, bridging design and code, or helping teams navigate change, we’d love to hear from you.

One thing that’s really important to us is leaving teams better than we found them. Not just with a cleaner component library, but with better habits, clearer processes, and the confidence to keep growing long after we’ve wrapped up.

If that sounds like the kind of work you enjoy, we’d love to hear from you.

➡️ Apply here: https://tally.so/r/2EZDK9