Christopher Patnoe

This year my main focus at Bonfire is building a new product on a very small team. I spend a lot of time actually coding, and it feels good. Being able to quickly deploy something without having to bother a fellow engineer has been a very liberating experience. More often than not, the amount of time to deploy the fix is less than it takes to document it, make a task, prioritize, and add to the backlog. I'm also not just talking about copy changes, or fixing padding values. Designers usually have a solid understanding of how the app is engineered, plus a deep knowledge of the the customer needs. Pair them with a LLM, and you can have a very powerful asset on your hands. For example, every week we have a design & UX review. A paint point came up last week—it wasn't clear how big the artwork should be when uploading it to a specific location on a product. Not knowing that info was a major paint point, since it was hard to prep a design file. I built a prototype an hour later. The right endpoint was already there, so all we needed was a function to calculate the right dimensions, and a new React component to display the guidelines. The next day it got merged to master and deployed. In a regular development cycle, this would have taken at least a week, and that's if you're lucky! If you're a designer, and you feel comfortable with frontend development, there's no reason to not have access to the production code of whatever app you're working on. I don't think it's a must on a bigger teams, since designers already have to wear many hats; but on smaller teams it can be a game-changer. It allows everyone to move faster; iterating and delivering new ideas quickly. It's 2025—let's normalize designers coding (and pushing to production). https://lnkd.in/exJa_CKi

32

8 Comments

Great talk from Luke-n-team at Zeroheight on MCP + AI and how tools like Cursor, Storybook, and Figma can connect into design system processes.

6

📣 Some good AI news for designers: agentic AI is just UX in disguise! You know when someone says something simple and suddenly the hard thing makes sense? Just had one of those 💡 lightbulb moments in a workshop with Chris Johnson about command line tools for process optimization with agentic AI. I've been watching a lot of videos about building AI workflows lately. About 30 seconds in, the presenter usually dives off into n8n or some other visual coding environment where they have 10000 boxes joined up with spaghetti string. If you just set all of this up, magic will happen! (Anyone who has used a prototyping tool has been there: no judgment!) CJ skipped past the complicated tools, and stuck with the concepts. Turns out that "agents" are just a very thorough and structured JTBD-like persona in a file that the AI uses as context for its work. Let's break it down a bit: ( 💻 AI-speak) -----------------------> (👩‍💻 UX-speak) As long as you have a tool capable of agentic AI (e.g. Claude), you create: * a "knowledge base/context" ------> "folder" of files, including * "skills" ------------------------------> "user expertise & knowledge" * "agents" ----------------------------> "persona" * "processes" -------------------------> "tasks, workflows & journey maps" ...to get things done. Who knew? UX-ers have been AI-ers all along! 🙌 I might just be the last person on the planet to "get" it (wouldn't be the first time...), but I figured maybe others have been as mystified as me. I'm pretty excited about the possibilities that this opens up: using a new technology to play a bigger UX game on our own home turf. Especially when you throw a little OOUX into the mix...😎 More from CJ here: https://lnkd.in/eunxQAXa https://lnkd.in/eFQds6KT https://lnkd.in/emvdvm5Q

2

2 Comments

Keeping cummings as my narrative North Star, AI tooling has gone from r-p-o-p-h-e-s-s-a-g-r to  ,grasshopper; in the last couple months, and as practitioners in industries driven by outcomes, we’re inescapably drawn to dwell on an absolutely wild range of implications.

2

How can you avoid the frustration of a diluted design vision? The design handoff process is one of the critical phases in product development. Keeping the design vision alive and intact until production is hard, and requires both excellent tools and deep understanding of the requirements. Jason Manns outlines both the recent history of design tools and the key elements that today help protect your original vision: 👇 https://lnkd.in/dfXMzT38 #uidesign #uxui #qt

12

Many design teams face the same tension: engineering moves very quickly, but good design exploration takes time. We often don’t have the luxury of spending weeks exploring and refining designs before implementation. So here at Uniphore we’ve been experimenting with how AI tools might help accelerate parts of the design workflow. One thing that’s becoming clearer is that not all features require the same design process anymore. We’re starting to think about two categories. **When designing for Pattern-based features** These are areas where the UI is mostly composed of existing components and patterns — dashboards, tables, configuration panels, workflow builders. For these, we’re exploring a faster pipeline: Problem definition → AI generates layouts using our component library → Frontend code generation → Engineering connects backend logic In this scenario, Figma becomes optional because the UI is assembled directly from production-ready components. But this only works if a few prerequisites exist: • A mature component library where visual design, states, and accessibility are well defined • A predictable layout system with primitives like grids, panels, and containers • A system-heavy product where the interface is structured UI rather than brand-heavy design When those conditions are in place, AI can assemble layouts that are already close to production. **When it comes to designing for exploration-heavy experience** These features require deeper design work — new workflows, interaction patterns, or complex user journeys. For these, the process still looks closer to traditional design: Problem definition → AI exploration for early directions → Figma refinement and prototyping → Engineering implementation The speed AI brings is real. But the bigger shift may be where we choose to design deeply. Instead of always designing screens, designers may increasingly focus on systems, patterns, and UX logic — while AI handles more of the UI assembly. AI isn’t removing design from the process. It’s changing where designers spend their time and what designers design. #ProductDesign #DesignSystems #AIinProduct #UXDesign #ProductDevelopment #DesignLeadership #AIAgents

9

A few weeks ago we released Gravity Sketch v6.5. As more design development happens in Gravity Sketch, we’ve seen design reviews expand beyond designers - to engineers, PMs, and product marketers. Conversations are not only about evaluating the aesthetics of the model but also about the functionality, ergonomics, and manufacturability. We started hearing questions like: ❓ "Will the vehicle harnessing fit under the hood?" ❓ "How comfortable is this power tool to grip?" ❓ "What space is left to enlarge the cab of this tractor, given the constraints of all the rotating parts?" v6.5 includes a few updates to support these cross-functional workflows: 1. Point-of-View lighting 🔦: Clear, consistent lighting for the user on desktop and in VR, regardless of scene lighting. This is critical for reviewing form and detail from every angle - this also fixes "the bottom of the shoe problem" (pictures attached) which shows an extreme scenario of poor lighting in action! We also provided much more control over the lighting and environment, so users can mimic the lighting of their design studios to evaluate models under familiar lighting and context 2. .blend export ⚙️: Designers and engineers often bring in engineering data from software such as Autodesk's Alias and Creo, a PTC Technology, build on top of it, and export it back out for refinement to add precision and detailing. A key software in this import/export loop is Blender. We now support native .blend export - so no need to reapply modifiers or clean up meshes 3. GS Labs 💡: Our new space for experimental features like speech-to-text and AI image-to-3D generation. We've had thousands of sign-ups so far, and as a result of feedback we moved speech-to-text into the app proper - so thank you to those who tried the features so far! A big thank you as well to the team in delivering this release. Also, we're cooking on a lot of exciting things at the moment so stay tuned! PS - thanks to our customers for sticking with the screen app even with all the lighting challenges 😅, looking forward to hearing all the other feedback now 👀 #IndustrialDesign #Design #DesignEngineering #3DModelling

76

5 Comments

🚨 How I Handled the Most Difficult Engineer I’ve Worked With 🚨 Let me break it down. He didn’t read the narrative. He didn’t prep. He didn’t know the workflows. And when I explained things? He gave me nothing but passive resistance and vague answers. Instead of throwing a fit, I ran the show. I treated the session like my demo. I guided the engineer click by click. I clarified workflows, redefined priorities, cut scope in real-time, and built training blueprints on the spot. This wasn’t just a tech walkthrough. It was a masterclass in leadership under pressure. Engineers, if you're invited to demo, come correct. Designers and IDs, stop waiting for ideal conditions. Own the outcome. 🧠 Lesson: Technical knowledge is power. But clear communication and ruthless focus? That’s how you win. #thejessemercado #LeadershipUnderFire #DesignOps #InstructionalDesign #NoExcuses

This is a great post from Sara Clayton and an important conversation as more designers start merging PRs with the help of AI agents. I'd add one thing: the hidden cost of speed. AI-generated code can accelerate velocity dramatically, but without discipline around design system constraints and engineering standards, it creates technical debt that quietly erases those gains. The companies getting this right treat design and engineering as co-owners of velocity.

9

1 Comment

My predictions for AI in 2026 are meant to help folks skip to the punchline before they have spent their budgets. These thoughts are informed by the experiences of the early adopters like the ones in the linked article. Predictions... 1) More executives become aware of what "probabilistic" means 2) Once they are aware of what that means they are going to get excited about RAG and adding guardrails 3) Once they realize what adding RAG and guardrails means from a cost, effort, and time point of view they are going to be a bit less excited about AI 4) Once they realize the impact RAG and guardrails have on the "magic" they were promised for their user experience they'll be even less excited https://lnkd.in/edNCrXYa

4

It's full steam ahead in engineering at Signapse. We recently launched #SignStream, a collection of APIs that enable real-time unconstrained translation of any sentence into sign language, but how does it really work? I've written a short blog that touches on how our API works and how you can get started.

8

If I have to explain a feature, the UX still needs work. Me understanding how it works does not matter. It also does not matter if it only makes sense after I have explained it. A feature proves itself when someone new looks at it and just gets it without extra explanation. That is why I value every “is this a bug?” question. If something looks like a bug, even when it is intentional, that is a sign the experience is still wrong. If users are confused, it is not their fault. It is ours. Features are not finished when they work technically. They are finished when they work intuitively.

11

9 Comments

Trust is the hardest metric to design for. Waymo just showed us how it’s done. Jonathan Slotkin’s recent analysis of Waymo’s 100 million driverless miles reveals a stunning UX stat: 91% fewer serious-injury crashes compared to humans, and 96% fewer intersection-related injuries compared with human drivers in the same cities. The scale of this gap reframes autonomous driving from a tech experiment into a public health opportunity. What stands out is the consistency of the results. Fewer high-impact collisions, fewer pedestrian injuries, fewer cyclist incidents. The pattern points to a system that follows rules, avoids distraction, and reacts faster than humans can. It solves the "Human Error" problem by redesigning the system, not the user. As UX professionals, we talk a lot about "human-centered design." Sometimes, that means acknowledging human limitations like distraction, emotion, and reaction time. The future of mobility and UX is about removing the burden of operation entirely. #ServiceDesign #UXResearch #Safety #AI #UX

9

2 Comments

The design-to-code handoff is one of the most expensive gaps in enterprise software development. Weeks of engineering time spent interpreting Figma files, reconciling spacing math, and rebuilding what designers already specified. On March 30, join Aditya Karra, Blitzy's Founding Designer, and Ben Bergerson, Senior AI Solutions Consultant, for a walk thru of Blitzy native Figma-to-production capability. No prototypes. No Refine PR. Production-ready code. - Every design token read before a single line of code is written - Responsive output across five viewports from one desktop frame - A new feature integrated into an existing codebase, brand-aligned, checked directly into the repo Designers and product teams can now move at the speed of design. Come see what that looks like in practice. Register with the link in the comments. March 30 | 12pm ET | Free with live Q&A #Blitzy #Webinar #Figma #Platform40 #FrontendDevelopment #EnterpriseAI #ProductVelocity

29

1 Comment

🌟 Google’s new #Stitch tool for Vibe designing is very impressive The Voice and the vision context together is powerful. It’s in Beta & you can try it here; https://lnkd.in/g75Z97YY Based on the recent developments in 2026, for me personally: - Anthropic is number one. - Google is number two. - OpenAI seems to be a distant number three, which is very surprising.

8

5 Comments