Ridd and Tommy spend two and a half hours on what Dive’s own talent network data actually shows: designers with 10+ years of experience account for about 45% of hires while 0 to 2 years accounts for about 7%, expected salaries at signup run around $200,000 against a bit over $215,000 actually paid, and the median from first meeting to hire is 38.4 days. The most uncomfortable stretch is the case that a FAANG résumé is no longer a positive signal on its own, with Decimals founder Danny confirming that people are ​“almost allergic to the big company logos.”

Marcin Wichary gave Unsung, his blog about software craft, its first proper design in time for the 500th post. He wanted something that ​“steered clear of the preeminent AI aesthetic,” and landed on a factory green he kept spotting in real places, ​“beautiful in its own pedestrian way.” Warbler for body text, Bild for the sans, a modified Neonoir for the logo, all of it holding the old tenets: no rounded corners, no shadows, no blurs, and a general distrust of symmetry.

Rogie King on how generative shaders went from a vibe-coded ​“FigGPU” prototype to the Config keynote in two months. The detail I keep coming back to: progressive blur took the Figma Draw team months to build, and Rogie recreated the same effect with one or two prompts. He’s also honest about what didn’t work – his working prototypes were great for designers and useless as references for engineers, PMs, and stakeholders, so the team backed up to static Figma mocks to get alignment first.

Anthony DiSpezio, Designer Advocate at Figma, builds a Make kit from an npm package. Make auto-generates a starting version of the guidelines, and Anthony offers a few rules for making them more effective – assume the agent knows nothing, split guidelines into categories so it doesn’t drown in context, include real usage examples, and spell out the gotchas. Before publishing your kit, calibrate against the model’s pre-training biases and test how it actually reads your UI.

Miggi rounds up ten community skills for the Figma agent, though the real news is in the first line: you can now author skills inside Figma. Ask the agent to draft one using a design frame as reference, preview it, then keep iterating in chat or open the markdown editor for manual changes. Emil Kowalski’s /find-animation-opportunities, which scans a file for places that don’t animate but should, and Jakub Krehel’s /better-interface which catches the details your interface is missing, are the ones I’d install first. 

The post above shows the skills, this one shows the plumbing. Miggi builds 3 skills: a type scale generator that also writes the variables, an image audit that drops a DPI card beside every image, and Shader Starter, which pairs neatly with the shaders update below.

Jake Albaugh, Developer Advocate at Figma, points a fleet of agents at an entire component library, then reuses the same agents as validators: each file re-fetches from Figma and surfaces what drifted after a property rename. His real argument for Code Connect is context economy – without it the MCP server hands the agent every style value it might need, and compaction is lossy.

“And suddenly, you’re just feeding the AI prompts. And eventually you’ve lost control. And you surrender to the AI.” Dylan Field on TBPN, right after Figma’s Q2 earnings, describing what CPO Yuhki Yamashita named quiet surrender. You start with a picture in your head, the model talks you into its direction, and you never notice the swap.

An hour with Miggi covering the full surface of Figma Motion: keyframes and eases, anchor points, easing and timing variables, loop tracks, motion paths, trim path on strokes, shader effects and fills, and a first look at 3D transforms. If the playlist above is the primer, this is the deep end, and there’s a community file to follow along in. One EDU caveat worth knowing: the Figma agent isn’t available on Education, so students work with the prebuilt shader presets instead of prompting their own.

Drew Wilson, founder of Opacity and the person behind ValioCon, is teasing another game that renders designers as horses, each with a custom neigh sound. I don’t want to spoil the end result, but please go check it out.

Same premise as Backyard Designers above, except Gabriel Valdivia built the whole game. Craftdeck is a playable CCG with mana, taunts, campaign bosses, and hundreds of designers pulled from real X profiles. Look for me as the Open Palette with a Version Control ability — this checks out.

Backyard Baseball, remade as an industry draft board. Ridd and Tommy Geoco put the design Twitter on a bleacher as cartoon kids, each with a scouting report, a portfolio link, and a little prop at their feet. You fill seven slots (a design leader, two product designers, a web designer, a design engineer, a brand designer, and a wildcard) under a limited budget, based on the follower count. The amount of fun little details is absolutely mind-blowing, and you’ll catch them only by exploring the community and drafting the dream team. Now, please excuse me as I go send some paper planes.

Luis Ouriach makes the provocative case that when LLMs make code cheap to write, edit, and migrate, the contribution model and the component-as-source-of-truth lose their reason to exist. His point is that consistency was always carried by tokens, not components, so a well-defined token set could govern ten different LLM-generated inputs at once. He stops short of full obituary, asking whether components live on as the unit of constraint.

Where the video above is Figma’s polished pitch, Mal shows the messy middle. She builds her own color-audit skill for the Figma agent, feeding it brand guidelines and example palettes so it flags off-brand elements by severity and annotates the exact violations right on the canvas, then has it rebuild the frame in every approved scheme. ​“Truly a diva just sharing her work in progress,” and better for it.

Until now, screens brought onto the canvas from code landed as raw values, leaving a cleanup pass before anything matched your design system. Figma now auto-binds colors, type, and spacing to the variables already in your file — whether you copy a Make preview, generate from code through the MCP server, or capture a live page with the Chrome extension. More frames also arrive with auto layout enabled, which dovetails with the Flexbox-mapped auto layout update above. Run Check Designs as a second pass to catch anything the matcher missed.

Miggi builds a loading spinner from one ellipse: swap the fill for a stroke with shift+X, sweep the arc tool to 75%, then round the end caps. In Figma Motion, the move is keyframing only the sweep value and deleting the start and ratio keyframes Motion adds by default, so it runs from 0.25 to a full 100%. Duplicate and stagger a few for a nested loader.

Miggi packs two Figma Motion tricks into one tip. Nest your animation inside a frame and the parent becomes a camera you can pan, zoom, and rotate. Separately, J and K jump the playhead to the next and previous visible keyframes.

New weekly live show by Ridd and Tommy Geoco, focused on design, tools, and the people behind the software.

Dylan Field responds to Gal Shir’s ​“quitting design” post that made a splash in the design community this week and supports findings in Lenny’s survey. He points to the recurring psychological loop that plays out every time a new AI model drops: existential crisis, experimentation, recalibration, repeat. His standing argument is that the attention economy makes design more valuable, not less, because anyone can prompt their way to average, and average doesn’t stand out. ​“This is the moment to be more bold, to take more creative risk, to double down on the power of design.”

Meng To sits down with Ridd at Dive Club for a 35-minute screen-share walkthrough of his current Codex workflow. He’s no longer opening Figma to create anything new — all design exploration, prototyping, and iteration happens in Codex, starting from screenshots and referencing his own past work as context. A few specific tactics stand out: using GPT Image 2 to generate UI concepts that then get turned into HTML, feeding a video of an animation to the AI and asking it to describe the right prompt to recreate it, and using design​.md files as a portable style guide that encodes years of taste into something the model can actually read.