My AI Grand Tour: Starting from NEAR Zero

The Game Plan

Keep myself motivated

This is technically Day 17. And there definitely has been frustrating days. But there has been and will also be those "Aha!" moments when something clicks. But until then I need to climb.

Don't wait for opportunities, take opportunities.

I'm not just winging it. I'm embarking on what I'm calling my "AI Grand Tour," a structured journey that starts with the fundamentals and builds methodically to expert level. The first stage is all about the bedrock: re-mastering computer science basics, getting truly comfortable with Python again, and—the most daunting part—learning the essential mathematics of Linear Algebra, Calculus, and Statistics. I never really liked maths, but I was good at it, and at an intermediate level of mathematics I should be able to re-learn what I did during University.

Maths in my opinion is the most boring part, but in the context of understanding Artificial Neural Networks maths is vital.

AI Co-Pilot not AI Autopilot

In order to learn I need to understand. Vibe Coding has become a very common phrase but expert level knowledge comes from being able to use the tools, and understand them.

Here's my twist. I am taking a build first approach and following these 3 paths:

1. Follow online courses that best target my initial AI aims. I've collated this list already.
2. Build Often and Build Fast
3. I'm going to learn AI by using AI. Not as a shortcut, but as a tutor.
4. Learn AI agent development and the tools that will make me a proper AI developer/engineer.

My rule is simple: the AI is my co-pilot, not my auto-pilot.

I'll use it to explain complex topics in simple terms, to help me find bugs in my code, and to suggest better ways of doing things. But I will not use a single line of code that I don't fundamentally understand. I'm learning to think, not just to type. This being the key mistake many people are making when writing hugely complex programmes and applications.

Everyone gets caught out with the Dunning-Kruger effect. My aim while learning Artificial Intelligence Development is to try and not get caught out, although I fear I ultimately will.

Why this matters

The thing is, I don't just want to be another person who can prompt ChatGPT well. I want to actually understand what's happening under the hood. When I build a neural network, I want to know why each layer exists, what the math is doing, how the gradients flow backwards. That's the difference between being an AI user and being an AI developer.

The tools I'm learning

Beyond the fundamentals, I'm diving deep into the actual tools that AI developers use every day. Things like building AI agents that can actually do useful work, not just chat. And getting my head around MCP (Model Context Protocol) - which is basically how different AI systems talk to each other and share information. These aren't just buzzwords, they're the building blocks of real AI applications.

Follow along if you want to see how this adventure unfolds. And if you're on a similar path, I'd love to hear about your experience. You'll find all my posts about learning in the title AI Grand Tour

Tags: #AI, #LearnToCode, #Python, #MachineLearning, #TechJourney, #AIGrandTour