Accessible only to conference ticket holders.
Log in Create account
For 90 days after a conference, only paid ticket holders can watch conference videos. After that, all Gold members have access.
If you have purchased recording access and cannot see the video, please contact support.
From prototype to production: Vibe coding design for real engineering systems
This video is featured in the AI and UX skill growth playlist.
Summary
Vibe coding can feel empowering for designers, but production code plays by different rules–and designers can’t see the full picture. In this panel, our panel members will unpack the hidden constraints, tradeoffs, and expectations that shape real codebases. Learn what engineers wish designers understood about AI‑generated code, and how to collaborate more effectively as design and engineering roles continue to blur.
Key Insights
-
•
Amelia's Intent workspace creates isolated, shared code environments that unify designers, developers, and product managers around the same product context.
-
•
Bringing designers into the engineering codebase requires specialized developer platform support to handle local runs and environment upkeep, as Elise notes.
-
•
Encoding broader design guidelines and semantics into code or accessible documentation enables AI agents and engineers to better understand and apply brand-consistent decisions.
-
•
AI tools excel at small, scoped tasks but struggle with global logic and large, complex problem alignment, as Amelia explains.
-
•
Collaboration challenges arise when designers new to codebases fear making mistakes; high-level education on engineering processes helps overcome this barrier.
-
•
David's team experiments with pod models pairing one engineer and one designer working closely to expedite design-to-code workflows while adjusting collaboration timing.
-
•
Role fluidity is increasing but must respect core domain expertise; designers typically focus on humane interfaces, engineers on architecture, even in AI-enhanced workflows.
-
•
Shared tooling and language around engineering basics (like Git) help non-engineers meaningfully participate in code-centric design processes.
-
•
Producing production-quality front-end code is becoming an expected skill for designers, moving towards a hybrid 'design engineer' role.
-
•
Successful AI adoption in design-engineering workflows requires planning and review phases before execution rather than relying solely on AI-generated output.
Notable Quotes
"It’s giving this kind of shared place where we’re all working, as opposed to previously designers are in Figma, devs in VS Code."
"Less handoff and more collaboration is becoming possible by bringing designers into the workflow with AI assistance."
"The engineering environment is a huge first hurdle — running and maintaining the local app can block designers."
"There’s so much context you have to add around your product because it might not even be named in the code base."
"AI models are good at local logic but not global logic, so keeping task size manageable is key."
"Spending more time upfront asking questions and planning before using AI agents to update code improves outcomes."
"We are rethinking pods with one engineer and one designer to promote more integrated product thinking early on."
"The job of design is not to become an engineer; raising the floor doesn’t mean ditching collaboration."
"There’s going to be role fluidity but also hardening of expertise because people care deeply about what they do."
"The artifact was never what created value — it’s the combined collaborative output that truly matters."
Or choose a question:
More Videos
"One theme was very clear and that was around the future."
Alana WashingtonTheme 3 Intro
October 1, 2021
"Pat Helisco funds projects that have social impact and provides designers space to exhibit and connect with stakeholders."
Sofía Delsordo Kassim VeraPublic Policy for Jalisco's Designers to Make Design Matter
December 8, 2021
"You have to put on many faces — different skills and mindsets — to communicate with purchasing, finance, legal, and HR."
Crystal PhilcoxThe Many Faces of Operations
November 6, 2017
"It is very difficult for somebody who has never used assistive technology to have the same experience as someone who uses it daily for everything."
Sam ProulxDesigning For Screen Readers: Understanding the Mental Models and Techniques of Real Users
December 10, 2021
"Data entry is the enemy; automation is the key to making research repositories successful."
Sofia QuinteroThe Product Philosophy Behind EnjoyHQ
March 10, 2021
"We make the invisible visible and turn abstract strategy into something people can see, touch, and act on."
Doug PowellDesignOps and the Next Frontier: Leading Through Unpredictable Change
September 11, 2025
"The most significant source of confusion is the difference between research questions and interview questions - Erica Hall."
Johanna KollmannInsights-Driven Product Strategy: Get your Research to Count
December 6, 2022
"ML ops offers governance, accountability, and a clear stakeholder responsibility framework borrowed from DevOps."
Joel BranchHumanizing AI: Filling the Gaps with Multi-faceted Research
March 11, 2021
"Creativity, accountability, and authenticity are going to be the new markers of humanity."
Katie JohnsonDisrupting generative AI products with just-in-time consumer insights
June 4, 2024
