The best way to make the modules stick is to build something. These six projects take you through the whole stack: deep learning in PyTorch, a Transformer built from scratch, a RAG pipeline, LangGraph agents, QLoRA fine-tuning and a production-grade MLOps workflow.
Each one is a code-along video, so you can pause, type it out yourself and adapt it to your own data. They are long, so pick the project that matches the module you've just finished and come back for the rest.
Go from PyTorch basics and linear regression to image classification, CNNs, residual networks, and training your own GAN, with notebooks to code along.
Andrej Karpathy builds a character-level Transformer language model from an empty file and trains it on Shakespeare, covering self-attention step by step.
Build a RAG pipeline from the ground up: indexing, retrieval, generation, then query translation (multi-query, RAG fusion, HyDE), routing, and adaptive RAG.
A hands-on course on graph-based agents: LangGraph fundamentals, five agents built from scratch, AI-agent design patterns, and adding RAG to a workflow.
Understand LoRA and QLoRA, then fine-tune Mistral-7B-Instruct into a custom YouTube-comment responder, with the Colab notebook and dataset provided.
A 3-hour project course using ZenML: set up the environment, design MLOps workflows, train and evaluate a customer-satisfaction model on the Olist dataset, then deploy it with a Streamlit app.