AI-Powered Frontend Development: Building Better UIs Faster

This ad is not shown to multipass and full ticket holders
React Summit
React Summit 2025
June 13 - 17, 2025
Amsterdam & Online
The biggest React conference worldwide
Learn More
In partnership with Focus Reactive
Upcoming event
React Summit 2025
React Summit 2025
June 13 - 17, 2025. Amsterdam & Online
Learn more
Bookmark
Rate this content

AI tools are changing how we design, build, and test user interfaces. This talk looks at current AI technologies that are improving the frontend development process and how to use them, including:

- Using Visual Copilot to convert Figma designs into production-ready code;

- Implementing AI agents for automated code writing and testing;

- Real-world examples showing how AI tools can improve development speed and joy.

This talk has been presented at Productivity Conf for Devs and Tech Leaders, check out the latest edition of this Tech Conference.

FAQ

Large language models (LLMs) are advanced predictive text systems powered by large neural networks, designed to predict the next word or text based on input. They mimic computational thinking and can act similarly to human brains if sufficiently large.

LLMs can enhance front-end development by automating code generation, iterating on code based on specific prompts, fixing errors, and providing design solutions that align with brand aesthetics.

Cursor IDE is a development environment that integrates with LLMs to automate and assist in coding tasks. It features agent mode for contextual code generation and error resolution without manual specifications.

Builder.io's Figma plugin uses AI to generate design elements that match brand styling. It allows users to create components like contact forms in Figma with AI-generated styling and functionality.

Agent mode in Cursor IDE is a feature that allows the IDE to automatically determine context, generate relevant code, and fix errors without the user needing to manually specify every detail.

Using tests with LLMs allows developers to write tests first, then code, letting the LLM iterate until the tests pass. This approach provides stronger guarantees of code correctness.

Builder.io's CLI can convert Figma designs to code directly in the IDE. It allows designs to be updated or modified in Figma and then automatically reflected in the codebase through simple CLI commands.

The author suggests using AI as a collaborative tool, similar to working with a junior developer. This involves providing feedback, refining processes, and using LLMs to automate repetitive tasks, enhancing productivity and creativity.

Coding is the primary use case for large language models. They are particularly effective in assisting with coding tasks, especially in front-end development.

Parallel development involves starting with draft designs and implementing functionality while allowing design changes to be integrated later. AI facilitates this by updating code to reflect design changes without requiring a complete overhaul.

Steve Sewell
Steve Sewell
19 min
27 Mar, 2025

Comments

Sign in or register to post your comment.
Video Summary and Transcription
Today's Talk introduces the use of large language models (LLMs) to enhance front-end development. LLMs can act like our brains by maximizing the good parts and minimizing the bad parts. A demo in Cursor, an IDE, showcases how LLMs can be used with the builder.io Figma plugin. The Talk emphasizes the automation of tasks, such as adding a settings button and resolving errors, with the AI agent. Feedback and manual verification are crucial to ensure desired results. Tests and continuous iteration are recommended for stronger guarantees of correctness. Monitoring and guiding the AI agents is important to stay on track. Connecting to other tools like Figma and using AI prompting can further enhance code generation. The CLI enables code base integration and parallel development. Visual prototyping and seamless updates are possible with the Builder tool. Overall, the Talk highlights how LLMs can revolutionize front-end development by automating tasks, improving efficiency, and facilitating collaboration.

1. Introduction

Short description:

Today I want to talk to you about how to build better front-ends faster by using LLMs. Large language models are just very fancy predictive text systems. They can start acting a lot like our brains think. The whole goal is to figure out how to maximize the good parts and minimize the bad parts. We'll focus on front-end development because LLMs are particularly strong there. I'm going to start demoing in Cursor, an IDE, and show you a basic use case with the builder.io Figma plugin.

What's up, everybody? Today I want to talk to you about how to build better front-ends faster by using LLMs in ways you probably don't already today. I guarantee at least one trick I show you here. So who am I? I'm the founder and CEO of Builder.io. I've also made some cool open-source projects. Let's talk a little bit more about some of this stuff in a bit.

So to set back, large language models, it's very easy to forget, are just very fancy predictive text systems. You give them a chunk of text and they predict what text would come next. That's it. All the amazing things really are just powered by that one basic technology. Now under the hood, there are very large neural networks. are designed after how our brains work. And it turns out, shockingly, that if you have a sufficiently large neural network, it can actually start acting a lot like our brains think. And it's almost like computational thinking, thinking on demand, which makes them surprisingly smart sometimes. And the thing that we tend to find with large language models is the whole goal is to figure out how to maximize the good parts and minimize the bad parts because there's lots of both. And interestingly, across LLM use cases, coding is the number one use case by most studies, especially when you look at what are people actually using and adopting most. So we have the privilege of being able to use this technology to help us with coding as it's surprisingly good at that compared to other tasks. And so we really want to dive into today a bit more on how can you best use them. We'll focus on front-end development in particular because LLMs are particularly strong there.

And from here forward, I will be live demoing and I'm going to start over here in Cursor. And let's go into this repo. Cursor is the IDE I'm going to be demoing on. Cursor is not the only great IDE that has these features. I would say WinSurf is another great option, cannot go wrong with it. But let's go into a basic use case here and let me explain first what I'm going to do. So I'm working on this Figma plugin, the builder.io Figma plugin. And I added a cool new feature where we can design things with AI. So maybe I'll choose like I want a contact form. And you can select something in Figma and it'll use AI to generate something based on what you ask for in the style of what you've selected. So that's a common problem with AI is I want not just generic Tailwinds looking AI slot, I want something that looks like my actual brand or the way I actually design, etc. Like here we can see it's building out my contact form, kind of step by step.

2. Adding a Settings Button

Short description:

I want to add a settings button that launches a pop over with a use auto layouts checkbox in my code base, without manually adding anything to context. The agent automatically figures out the relevant files and context.

And it looks with similar styling, similar branding as what you saw. And then when I'm ready, I can import it into Figma and then ba-bam, we've got it. Great. But I want to add a new feature to this. And the feature involves a UI, a button to pop over a few things. So I want to add like a settings button next to this button so that I can change what style of import. And I do. Right now we just do a generic import, but I want to make it so that we can also import using nested frames and auto layout. And so whereas you'd normally go and start building out the code for all this, let me show you a much easier way that I use pretty much every day.

Jumping back over to my IDE, I'm going to go over to cursor and use their agent mode. This is their new UI, by the way. I think most people have the new UI, but agent mode, I believe is the default now. You can also get to a command I, pop it in and out. So here I'm just going to type what I want. Now this is a real full-size code base. This code base has a lot of files, a lot of stuff in it, a lot of code. And what I'm going to do is I'm going to not add anything manually to context. I'm going to let the AI figure things out. So I'm just going to say next to the import to Figma button, let's add a settings button that launches a pop over that has a use auto layouts checkbox. When that box is checked, set use frames to true in HTML2 Figma. And let's go. I'm getting pretty specific there with my prompts and I'm just telling it like, Hey, I want to add, let's go up again, saying, Hey, I want to add a settings button. I didn't describe how I want it. Generally these agentic features are pretty good at figuring out use a certain icon library, I'm going to use that icon library, et cetera. I want to launch a pop over and what the behavior should be. And the agent automatically can start figuring out what files are relevant here. So it did a few kind of lookups here. I don't know why it's hitting some random errors, but whatever. But it's able to figure out the context automatically. I didn't have to specify it myself.

Check out more articles and videos

We constantly think of articles and videos that might spark Git people interest / skill us up or help building a stellar career

AI and Web Development: Hype or Reality
JSNation 2023JSNation 2023
24 min
AI and Web Development: Hype or Reality
Top Content
This talk explores the use of AI in web development, including tools like GitHub Copilot and Fig for CLI commands. AI can generate boilerplate code, provide context-aware solutions, and generate dummy data. It can also assist with CSS selectors and regexes, and be integrated into applications. AI is used to enhance the podcast experience by transcribing episodes and providing JSON data. The talk also discusses formatting AI output, crafting requests, and analyzing embeddings for similarity.
Less Cruft, More Power: Leverage the Power of the Web Platform
C3 Dev Festival 2024C3 Dev Festival 2024
30 min
Less Cruft, More Power: Leverage the Power of the Web Platform
This talk focuses on the powerful features of CSS and HTML that can level up developer skills and enhance web UI. Scroll-driven animations, popover API, and anchor positioning are explored as ways to create dynamic effects, improve performance, and increase accessibility. The talk also emphasizes the benefits of building presentations with CSS and HTML, separating logic from styling, and leveraging core platform features. Wishlist features for the web platform and the challenges of pushing updates across browsers are discussed.
Forget Bad Code, Focus on the System
React Summit US 2023React Summit US 2023
27 min
Forget Bad Code, Focus on the System
Top Content
Watch video: Forget Bad Code, Focus on the System
Setting up the system and separating concerns are important in software development. Modular construction and prefab units are a new trend that makes construction quicker and easier. Architectural complexity can lead to a drop in productivity and an increase in defects. Measuring architectural complexity can help identify natural modules in the code. Best practices for avoiding architectural complexity include organizing code by business domain and using prop drilling. Atomic design and organizing a monorepo are recommended approaches for managing architectural complexity.
Confessions from an Impostor
JSNation 2022JSNation 2022
46 min
Confessions from an Impostor
Top Content
The Talk discusses imposter syndrome and reframes it as being a professional imposter. It emphasizes the importance of sharing and starting, embracing imposterism, and building inclusively for the web. The speaker shares personal experiences of being an imposter in various technical disciplines and highlights the significance of accessibility. The Talk concludes with the idea of building a collective RPG game to remove excuses for not making things accessible.
Improving Developer Happiness with AI
React Summit 2023React Summit 2023
29 min
Improving Developer Happiness with AI
Watch video: Improving Developer Happiness with AI
GitHub Copilot is an auto-completion tool that provides suggestions based on context. Research has shown that developers using Copilot feel less frustrated, spend less time searching externally, and experience less mental effort on repetitive tasks. Copilot can generate code for various tasks, including adding modals, testing, and refactoring. It is a useful tool for improving productivity and saving time, especially for junior developers and those working in unfamiliar domains. Security concerns have been addressed with optional data sharing and different versions for individuals and businesses.
Maximize Productivity with AI Agents
Productivity Conf for Devs and Tech LeadersProductivity Conf for Devs and Tech Leaders
25 min
Maximize Productivity with AI Agents
I'm Tejas Kumar, a software developer with over 20 years of experience. AI agents are defined as entities that act on behalf of users or groups to produce specific effects. Agents consist of an orchestration layer, a language model, and tools represented as JSON functions. Langflow is an open-source tool that allows users to build their own AI agents by connecting language models and tools. Composio is a tool that enhances agent capabilities by offering integrations and apps, such as Google Calendar integration. MCP (Model Context Protocol) is a way to share context with models and extend their capabilities. It allows functions to be made available to models over standard input/output or an HTTP endpoint. MCP can be used with GitHub to perform various tasks like searching and fixing code issues. The Talk covered the basics of AI agents, building agents with Langflow and enhancing them with Composio, and using MCP with GitHub. The speaker encouraged audience questions and exploration of these concepts.

Workshops on related topic

How to Create a Web Application in an (Almost) Autonomous Way Using Clean Coder
Productivity Conf for Devs and Tech LeadersProductivity Conf for Devs and Tech Leaders
95 min
How to Create a Web Application in an (Almost) Autonomous Way Using Clean Coder
Workshop
Grigorij Dudnik
Grigorij Dudnik
Imagine replacing yourself with a multi-agent AI programmer to develop your production web application. That's exactly what we did at my startup takzyli.pl. To achieve this, we designed and used the Clean Coder - AI agent framework for autonomous code writing (https://github.com/GregorD1A1/Clean-Coder-AI), which is hopefully open-source project. If it worked for us, why shouldn't it work for you?In this workshop, I'll show you how to create an entire web application in an (almost) autonomous way and drastically reduce the time you or your employees spend on writing code.
Test, Code, Repeat: Mastering AI-Assisted Development
Productivity Conf for Devs and Tech LeadersProductivity Conf for Devs and Tech Leaders
53 min
Test, Code, Repeat: Mastering AI-Assisted Development
Workshop
Marco Pierobon
Marco Pierobon
"Test, Code, Repeat: Master AI-Assisted Development" introduces developers to a transformative way of coding with AI as a collaborative partner. This workshop focuses on how iterative workflows, such as the ping pong pairing technique, enable an enhanced interaction between human creativity and AI efficiency.