This blog already runs four series in parallel — VicinoTe, AI with Ruby, Ruby Deep Dive, and WhatsApp with Rails. Two more are joining soon, and this post is a short heads-up on both before the first real episode of either lands.

Semantic SEO: Beyond the Keyword

For a long time, ranking well in search meant matching the right keywords. That’s not really how modern search works anymore — search engines increasingly rank on meaning, not just on which words appear on a page. This series is about that shift: what “semantic” search actually looks at, why keyword-stuffing stopped working, and what it means, practically, to write and structure content for a search engine that’s trying to understand intent rather than count occurrences.

This one is deliberately not Rails-specific or code-heavy — it’s for anyone who writes online, runs a website, or is just curious how search really works today.

Semantic Search in Ruby on Rails

This series is the technical counterpart: what semantic search actually is under the hood — embeddings, vector similarity, nearest-neighbor lookups — and how to build it into a real Rails application from the ground up. AI with Ruby already gave semantic search one episode, wiring pgvector into a RubyLLM chat history; this series goes further and deeper, on its own, as the concept deserves.

Follow along

Neither series has a fixed start date yet — like Ruby Deep Dive and WhatsApp with Rails, they’ll be written alongside the ongoing rotation whenever there’s time to do the topic justice. Follow the blog to catch the first episode of each as soon as it’s up.