What happens when you need to refactor 377 files in a SvelteKit codebase? Without AI, you retire. With Claude Code, you ship it in 11 hours.

This talk walks through real-world examples of using Claude Code for massive refactoring work that would be soul-crushing to do manually. I'll cover:

• How I approach large-scale refactors with Claude (377 files → 364 files → and counting)

• Building and activating Claude Code skills reliably (based on my viral post that's getting 4k+ views in 3 days)

• The workflow patterns that actually work vs the ones that waste your time

• Real examples from production codebases

This isn't about AI replacing developers - it's about using AI to handle the tedious parts so you can focus on the interesting problems.

Since OpenAI released ChatGPT in late 2022 we've seen a lot of very interesting things pop up in the AI space. Among them are things like Vercel's v0 and openv0. In this talk we'll take a look at how to leverage these tools to build Svelte components.

Problem: Traditional AI application development often involves juggling multiple tools, APIs, and frameworks, leading to inefficiencies, complexity, and limited scalability. Developers are stuck in outdated paradigms that struggle to keep up with the growing demands of modern AI solutions.

Solution: The Model Context Protocol (MCP) offers a transformative approach to AI app development—essentially function calling on steroids. MCP represents a paradigm shift by streamlining the integration and management of core functionalities like file systems, databases, and other essential services through its innovative framework. By enabling seamless connections and robust functionality, MCP revolutionizes how developers build and deploy smarter AI-driven applications.

Description: This hands-on workshop empowers JavaScript developers to explore all facets of MCP development. Participants will learn to build MCP clients, set up MCP servers, and leverage existing MCP servers to create robust AI applications. Through practical exercises, you'll discover how MCP simplifies development, boosts efficiency, and delivers smarter AI-powered solutions. You'll also gain insight into how MCP's groundbreaking function-calling model enhances workflows, allowing seamless management of tasks such as file systems, databases, and more.

Key Takeaways: - Master the fundamentals of MCP and its innovative function-calling approach. - Build and configure MCP clients and servers from scratch. - Explore how to utilize existing MCP servers to streamline development tasks. - Gain hands-on experience in transforming traditional AI workflows into scalable, efficient models.

Abstract

Libraries and frameworks are where LLMs often break down: too much context, too many moving parts, and lots of hidden assumptions. In this talk, we’ll unpack why models struggle in this space and present our method for structuring library knowledge into digestible chunks. Using Svelte as the running example, you’ll see what goes wrong when models get it wrong, and how Tessl's approach can help them finally get it right.

Overview

Why frameworks and libraries trip up LLMs

Software engineering, among other things, requires precision in language, API patterns, and dependency versioning. But models are trained on snapshots of data that:

• Are taken in the past, meaning they have no knowledge of a framework just released;
• Contain mixed and sometimes contradictory information about the same package. For example, different versions without clear differentiation;
• Provide an unbalanced distribution of information. Some packages are well-documented, while others have very little coverage.

We’ll unpack why this matters for developers relying on AI coding tools and introduce a practical tool to help.

Tessl's method

• Structured, doc-like blueprints of a package’s API, best practices, and examples.
• They give coding agents a reliable reference point while iterating, helping them stay aligned with how a library is really meant to be used.

Svelte as a running example

• Svelte poses a particular challenge: LLMs often mix it up with other frameworks or fall back to using older Svelte versions patterns.
• With our method, the same agent can navigate the framework more effectively, producing cleaner, more accurate code.