From PubMed to Instagram: my first n8n automation
The first no-code automation I built with n8n: I give it a topic, and using Google Gemini the system searches PubMed for scientific papers, picks the best ones, and generates Instagram carousels (slides) from them. My goal was to automate content production for the gym.
How it works
What it produces
Pick a topic and browse the kind of carousel the automation produces.
Creatine monohydrate
One of the most researched sports supplements, with some of the strongest evidence behind it.
Sources (PubMed): PMID 28615996 · PMID 33557850
Generated from PubMed papers with Gemini
Why n8n
After discovering that backend existed, I got curious and started researching on YouTube. Burhan Kocabıyık's videos introduced me to an approach: instead of learning Python or JavaScript, learn n8n and build the systems you need far faster. The idea of creating something from nothing really drew me in.
What I built
My first goal was to solve my own work: producing content for the gym. I'd give the automation a topic; Gemini generated a PubMed query for it, the system searched and fetched the papers, selected the most relevant ones, and then Gemini wrote Instagram carousel (slide) copy from them. That turned science-backed content into an end-to-end automated flow.
What I learned
This project taught me the behind-the-scenes work at a foundational level. For the first time I saw API logic: how to send a request to a service (PubMed), how the returned response (XML) is processed and turned into JSON, and how data is passed between steps and filtered. Even though n8n is no-code, I grasped how a backend actually works and how data flows end to end. I saw the power of learning by solving my own need; this project opened the door to all my later automation work.