Start Here
What AI engineering actually involves, what this track covers, and the honest ratio of glamour to plumbing.
What this track is
AI engineering means building software where part of the behaviour comes from a model instead of an if statement. A model is a function whose rules were learned from data, not written by hand.
You will work through five stages:
- Foundations: Python, NumPy (the array library), and just enough linear algebra, calculus and probability.
- Machine learning: regression, classification, evaluation, overfitting, classic models.
- Deep learning: neural networks, backpropagation, training, PyTorch (a deep learning library).
- Transformers and LLMs: attention, tokenization, training, sampling. An LLM (large language model) is a neural network trained to predict the next piece of text.
- Building AI apps: APIs, RAG, agents, evals, fine-tuning, serving, safety.
1 of 8