Copy the formatted transcript to paste into ChatGPT or Claude for analysis
Hi there, I'm Gus Pelogia, SEO product manager, and I'm here to talk to you about
vibe coding your own SEO tool.
Now let's just get straight out of the bat.
I'm not promising that you're going to build the next revolutionary AI SaaS and you're
going to make a lot of money.
This is about making little tools that will save you time on your day-to-day, maybe get
you away from a bunch of boring tasks and then just speed up the process, maybe give
you a good draft of things that you can review yourself and then make it your own.
So there are three things that you need to vibe code your own tools.
First you need an LLM.
I start with ChatTPT, you might try another one.
You may need an API to connect the dots between your LLM and the place where your code's going
to run, and your code can be on Google Colab, which is a Python environment that runs on
your browser.
You don't have to install anything.
AI-generated overview
Gus Pelogia demonstrates "vibe coding"—using LLMs like ChatGPT to generate functional SEO automation tools in 15-30 minutes without traditional programming skills. The workflow involves prompting ChatGPT with specific requirements (Google Colab or Sheets format, API calls, CSV input/output), then copy-pasting generated code into execution environments. Core concepts include vector embeddings (transforming page content into numerical representations) and cosine similarity (matching content based on semantic closeness). Practical applications include tag-to-page matching for CTAs, entity confidence tracking via Google Knowledge Graph API, automated hreflang suggestions across languages, content decay analysis comparing historical traffic, and related page discovery. Pelogia emphasizes this approach saves time on "manual labor" data crunching, freeing SEOs to focus on strategic decisions. The process requires an LLM, relevant APIs (OpenAI for embeddings, Google Knowledge Graph for entity data), and an execution environment like Google Colab or Sheets.
"Vibe coding" allows SEOs to build functional automation tools in 15-30 minutes by prompting ChatGPT to generate code for Google Colab or Sheets, eliminating the need for traditional programming expertise.
Vector embeddings transform entire page content into numerical "corpus" representations, enabling semantic matching that works across languages—demonstrated with successful hreflang matching between pages in multiple languages.
The prompt formula is: (1) specify platform (Google Colab/Sheets), (2) declare API calls needed (e.g., OpenAI API), (3) request CSV input/output format, (4) explicitly define column contents (URL in column A, embeddings in B, tags in C).
Entity confidence tracking via Google Knowledge Graph API can run automatically on Google Sheets daily, monitoring brand entity strength without manual checks—Pelogia's personal name tracker has run daily for a year.
Want to build custom SEO tools without being a full-stack developer? In this Whiteboard Friday, SEO Product Manager Gus Pelogia explains the concept of "vibe coding"—using AI to generate code for simple, time-saving scripts that automate daily tasks. Gus walks through essential steps, including using an LLM like ChatGPT, connecting APIs, and running code in browser-based environments like Google Colab or Google Sheets. He also shares specific prompting strategies to ensure your AI-generated code works correctly. Download the HD whiteboard: https://moz.com/blog/vibe-coding-seo-tools-whiteboard-friday TIMESTAMPS: 00:00:00 – Intro: What is "Vibe Coding" for SEO? 00:00:39 – The 3 Essentials: LLMs, APIs, and Google Colab 00:01:22 – How to Prompt: Rules for generating working code 00:02:45 – Core Concepts: Vector Embeddings & Cosine Similarity 00:03:47 – Using Google Knowledge Graph API 00:05:13 – Tool Idea #1: Tag Matching 00:05:44 – Tool Idea #2: Entity Confidence Tracker 00:06:2