As a UX Lead at nvisia, I’ve spent the past few years navigating the intersection of design, technology, and team workflow. Like many in our field, I initially met the arrival of AI with skepticism—uncertainty about its impact, concern about creative devaluation, and a lot of open questions. But over time, that skepticism has transformed into curiosity, and ultimately, into practical excitement.
Because here’s the truth: AI isn’t here to replace UX designers—it’s here to empower us.
What we’re witnessing isn’t automation in the traditional sense. It’s a fundamental shift in how we work—one that opens doors to faster iteration, deeper user insights, and a more strategic role for designers as systems architects and experience stewards. And it’s all being accelerated by one critical evolution: the rise of localized AI workflows.
The AI Prototyping Engine: Speed & Exploration
The reality is this: AI isn't stealing jobs; it's offering us a powerful new tool. Think of it as a rapid prototyping engine – capable of generating diverse design variations at unprecedented speed. Imagine needing several landing page layout options quickly, or exploring countless button styles and icon sets without the tedious manual work.
AI can generate low-fidelity prototypes to test core interaction flows in minutes, dramatically accelerating the initial exploration phase.
It’s not about blindly accepting those outputs; it's about leveraging them as starting points—a springboard for further refinement. For example, a designer might input “e-commerce product listing – clean, modern aesthetic” into an AI tool running locally via LM Studio, and instantly receive five distinct layout proposals.
This accelerates the process, but remember: AI-generated outputs are starting points, demanding human curation, critical evaluation, and meticulous refinement.
Setting Up Your Local AI Workflow
The key to repeatable results lies in establishing a robust local environment. We’re advocating for a workflow centered around LM Studio, which lets you download and run powerful Large Language Models (LLMs) directly on your machine, bypassing cloud-based services entirely.
This offers significant advantages: enhanced privacy, reduced latency, and complete control over the AI model used.
Complementing LM Studio is VS Code with Cline. Cline acts as a command-line interface for interacting with LLMs, streamlining prompts, managing API keys (even when running locally), and integrating AI workflows directly into your development environment.
We’ve found Claude 3 Opus to be particularly effective for design tasks due to its superior reasoning capabilities and creative output – easily accessible through Cline within VS Code.
The Evolving Role of the UX Designer
The ability to run LLMs locally with LM Studio and manage them efficiently through VS Code + Cline empowers designers to experiment rapidly and iterate consistently. This isn’t just about efficiency; it's about reclaiming ownership of the design process.
Consider designing an e-commerce checkout flow: the designer uses a carefully crafted prompt within Cline to generate layout options via Claude 3 Opus running locally.
The designer selects the most promising one, refines it based on usability testing, ensures it aligns with brand guidelines, and then presents it to stakeholders, justifying its design choices and highlighting user benefits – all documented within the VS Code environment for future reference.
The Collaborative Workflow: A Continuous Cycle
The most effective approach isn’t about replacing human input; it's about fostering a collaborative cycle. Imagine this workflow: The UX designer crafts a detailed prompt for the AI, outlining desired functionality and aesthetic direction – this prompt is saved as a reusable template within VS Code using Cline.
The AI rapidly generates prototype variations. The designer then meticulously evaluates these prototypes, identifying strengths and weaknesses based on user feedback and design principles.
This refined information feeds back into the AI, prompting further iterations—a continuous loop of innovation. Clear communication between designers and any AI specialists involved is vital for maximizing this synergy.
Supercharging UXR with Machine Learning
The potential extends beyond prototyping. Machine learning algorithms can analyze vast datasets of user behavior, identifying patterns and predicting potential usability issues before traditional testing even begins. Imagine anticipating where users might struggle based on aggregated data—allowing you to proactively address those pain points in the design.
Personalized research methodologies, tailored to individual user profiles and preferences, are also becoming increasingly feasible. However, ethical considerations surrounding data privacy and responsible AI usage within UXR remain paramount – transparency and respecting user consent aren’t optional.
The Future of Design?: Scalable Experiences, Human-Centered Design
The future of design isn't about humans versus machines; it's about humans and machines working in harmony – especially when those machines are running locally under our direct control. We envision a world where robust design systems power seamless user experiences across platforms, AI augments human creativity, and UX designers lead the charge in ethical and inclusive design practices.
Embrace these changes, experiment with tools like LM Studio and VS Code + Cline, and advocate for a human-centered approach to AI integration.
AI isn't a threat; it’s an opportunity—a powerful tool that can empower UX designers to create even more impactful and meaningful experiences.
Originally published on nvisionaries.
Meet the Expert: Mike Arce
Human-Centered Technologist Reimagining UX Through Local AI Workflows
Mike Arce, UX Lead at nvisia’s Milwaukee Region, is one of the rare designers who thrives at the intersection of experimentation, systems thinking, and practical curiosity. His journey into AI-powered UX design started as soon as GPT-3 hit the scene. Not content to read about it, Mike jumped in—prompting models to describe interfaces, generate code, and testing the results directly in Codepen. This wasn’t just tech play—it was about closing a persistent gap in UX: the often messy handoff between design and development.
For Mike, that pain point has always been personal. Long feedback loops, mismatched interpretations, and designs getting rewritten due to engineering constraints all motivated him to seek faster, cleaner ways to move from concept to reality. But as more AI design tools like v0.dev and vibe emerged, he noticed a shift—some tools weren’t just augmenting designers, they were attempting to replace them. That realization sparked a deeper shift in his focus: toward localized, open-source AI workflows that give designers full control over the process.
Mike now champions a setup that includes LM Studio, VS Code, Cline, and Claude 3 Opus—a local AI toolchain that’s fast, private, and adaptable. But more importantly, it’s rooted in the values he holds as a UX designer: clear thinking, human-centered experimentation, and creative ownership. He’s gone beyond using tools—he’s trained models, tested agentic systems, and explored peer-to-peer research assistants, all to better understand the world users are about to enter as AI becomes embedded in digital experiences.
He believes the UX role is changing—not in essence, but in scale. Designers still need to own their design systems and make thoughtful choices for users. But with AI, we can now move at the speed of insight. What once took weeks of research and iteration can happen in days—or even hours. The key, Mike emphasizes, is that designers must remain active shapers of the process, not passive recipients of whatever the tool suggests.
Looking ahead, Mike sees both opportunity and risk. The opportunity lies in helping ideas become real—faster, smarter, and more securely. The risk? Relying too heavily on AI outputs without asking why they work, or if they even should. That’s why his mantra to fellow designers is clear:
“Start small, experiment widely—and stay grounded in what makes us human.”
Because for Mike, the future of design isn’t AI vs. human. It’s a deeper collaboration—where designers lead with empathy, and AI becomes a powerful, flexible tool in service of better experiences.