After years as an AI/ML engineer, writing books on ML and GenAI, and mentoring thousands of students, I’ve noticed the real issue isn’t too few tutorials; it’s that there’s not enough curation. If you want to move past endless tutorials and start building real-world projects, you need the right AI learning resources. Look for materials that focus on solving problems and practical engineering, not just theory.
Today, I’m sharing a carefully chosen list of the best resources I recommend to learners. These aren’t just links to save and forget; they’re the tools and courses that will help you go from understanding what an LLM is to actually deploying one.
The 5 AI Learning Resources You Actually Need
Here’s my go-to list of resources that cover all areas of modern AI engineering.
1. Google’s Introduction to Generative AI (Free)
Before you start working with complex prompts or building AI agents, it’s important to understand how Generative AI works.
Google Cloud has a great, free Introduction to Generative AI micro-learning course. It’s short, straightforward, and avoids the usual industry hype.
I always recommend beginners start here because it teaches you the right vocabulary. Whether you’re in an interview or pitching a project, you’ll need to explain why a Generative AI solution fits better than a standard predictive ML model. This course helps you do that. Find it here.
2. The OpenAI Cookbook
If you keep one tab open while coding, make it the OpenAI Cookbook.
Unlike regular API docs that only explain endpoints, the cookbook gives you real examples. It’s full of Jupyter notebooks and Python scripts that show you how to handle real-world tasks.
Recently, I deployed a RAG-based (Retrieval-Augmented Generation) app on a Linux server for a client. The theory behind RAG is simple, but putting it into practice can get tricky. The OpenAI Cookbook has clear code examples for setting up vector databases and semantic search with embeddings. It saves you hours of debugging and shows you how to structure your code the way professionals do.
3. ChatGPT Prompt Engineering for Developers (Coursera)
Prompt engineering isn’t just about typing well into a chat box; it’s a real engineering skill. The ChatGPT Prompt Engineering for Developers course on Coursera is a must-take.
Think of this as a coding course, not a writing class. You’ll learn how to use the OpenAI API to summarize, infer, transform, and expand data with code.
This course shows you how to build software where the LLM is just one part of the system. It’s very useful for data engineering and automation. If you want to create custom chatbots or turn messy, unstructured data into clean JSON, you’ll use these techniques in real projects.
Looking for a practical guide to GenAI? My book Hands-On GenAI, LLMs & AI Agents teaches LLMs, RAG, and AI Agents through real-world projects.
4. Andrej Karpathy’s “Neural Networks: Zero to Hero” (GitHub Repo)
I strongly recommend building with APIs, but at some point, you need to understand what’s happening inside the black box. The GitHub repo karpathy/nn-zero-to-hero is a top resource for this.
This video series and code repo features Andrej Karpathy (a founding member of OpenAI) building neural networks from scratch, step by step, all the way up to modern transformer models.
This is an advanced resource. Don’t start here if you’re new, but once you’re comfortable with APIs, this repo will help you level up. Knowing the math and code behind backpropagation and transformers makes you much better at debugging when your AI models start to hallucinate or fail in production.
5. My List of 40 AI/ML Projects
Theory doesn’t matter if it doesn’t lead to a working project. I often get messages from learners asking, “What should I build?” That’s why I put together a list of 40 AI/ML projects to help you step out of your comfort zone.
Choose a project that interests you. Don’t just copy and paste the solution. Read the goal, try to build it yourself with help from the OpenAI Cookbook and docs, and only check the solution if you’re really stuck.
I made this list to reflect real industry needs. These are the kinds of projects that stand out on a resume. Hiring managers aren’t interested in tutorials; they want to see that you can solve problems from start to finish.
Summary
Here are the AI learning resources I recommend you bookmark:
- Google’s Introduction to Generative AI
- The OpenAI Cookbook
- ChatGPT Prompt Engineering for Developers
- Andrej Karpathy’s “Neural Networks: Zero to Hero”
- My List of 40 AI/ML Projects
I hope you found this article on AI learning resources helpful.
For more AI and machine learning tips, follow me on Instagram. My book, Hands-On GenAI, LLMs & AI Agents, can also help you advance your AI career.
