Hands on workshop

This workshop will focus on addressing website performance issues with the help of Large Language Models, or LLMs, frequently called “AI”.

Participants will bring a laptop and a site they have access to making changes they want to work on performance for. First we will explore some ways that participants can use AI/LLMs to investigate their performance problems and recommend fixes as well as testing strategies.

Then we will each try a series of prompts against data about our sites, asking the model to interpret the data and correlate problems to specific plugins or code.

This is an experimental workshop The goal is to learn more about using AI as a performance assistant, a somewhat different use case typically associated with AI. Results will be unpredictable!

Required to participate

Participants should be comfortable building websites and familiar with performance metrics like Time to First Byte and Largest Contentful Paint.

Participants must have an “paid” account set up with Chat GPT (Plus), Gemini (Pro) or have some other LLM access that accepts file uploads as we will be uploading Lighthouse report JSON, webpage HTML, screenshots, Analytics data, etc.  Both of the paid services above (ChatGPT/Gemini) allow canceling at any time, so participants set up before the workshop and cancel after.

Takeaways

Here are some potential takeaways:

Understanding AI’s role in website performance optimization: Participants will learn how AI can be used to identify performance bottlenecks and suggest improvements.

Hands-on experience with AI tools: Participants will gain practical experience using AI platforms to analyze disparate data and generate actionable insights.

Knowledge of performance testing strategies: Participants will learn how to effectively test website performance AI-driven techniques.

Insights into AI’s potential for performance assistance:  Participants will explore a novel use case for AI, expanding their understanding of its capabilities beyond content generation.

Here are some expected achievements:

• Identify a performance bottleneck
• Implement an AI suggested fix