If you had to answer the question, "What is the mother of learning?" what would you say?
It is a slightly philosophical question, and the answer varies. Some might say patience. Others might say curiosity. For me, the absolute mother of learning is applying.
(Quick aside for the fantasy fans: Mother of Learning also happens to be the title of an incredible, completed web novel where the protagonist masters magic by reliving a one-month time loop over and over. Highly recommended!)
Universally, the core of human skill acquisition is progressive repetition. We try something, we fail, we adjust our approach, and we try again. We loop through the problem until it clicks.
The most fascinating part about this concept? That is exactly how Artificial Intelligence learns, too.
The Human and Machine Parallel
When we look at Machine Learning, the models go through "epochs" — massive cycles of progressive repetition. They ingest data, make a prediction, calculate the error, adjust their parameters, and repeat.
Whether human or machine, mastering a new skill requires looping through the data until the patterns make sense.
My AI Journey: From Curiosity to Production
My own journey into the world of AI mirrored this exact process. Back in 2022, I didn't start with complex architectures or multi-tenant databases. I started simply as a curious user messing around in the OpenAI Playground.
I was hooked. I spent my weekends reading articles, consuming documentation, and testing prompts. Soon, I was in the Playground at least five times a week, pushing the limits of what the models could generate.
But I realized early on — whether I was 13 writing my first lines of QBasic, or later building messy full-stack applications to connect NGOs with volunteers — that a shipped, imperfect product teaches you more than a polished tutorial ever could.
So, I started building. I used progressive repetition to go from testing basic prompts to engineering actual products:
- Integrating OpenAI's Realtime API and WebRTC to build voice-based interview simulators like InterView.ai.
- Building custom generative engines and leveraging the Gemini API for interactive storytelling platforms like FableWeaver.ai.
Architecting the Future
Fast forward to today, and that initial curiosity has evolved into my daily reality as an AI Engineer.
I went from just testing prompts to building entire platforms from scratch. Today, my work revolves around architecting full-stack systems with Next.js, TypeScript, and AWS. I build platforms where AI handles the operational grunt work so humans can focus on the work that requires actual judgment.
The Takeaway
If you want to understand Artificial Intelligence, you cannot just read about it. You have to get your hands dirty.
Find a tool, hit the API rate limit, figure out why your WebSocket dropped, and build it better.
The mother of learning is applying. So, what are you building today?