A free course from EurekaOh
Grounded AI
Understand the ideas behind artificial intelligence.
Start with an everyday example. Follow it through the mathematics, then see it work in Python. Build your understanding from your first model to modern AI.
We're preparing the first lessons. Each published lesson will have a written explanation, Python examples, practice, and a companion YouTube video.
Explore the curriculumNew to AI? You're in the right place.
This course is for AI beginners and people who already code and want to understand AI. We introduce the mathematics as it becomes useful. Python is our language for examples: we explain what the code does and how it connects to each idea, without teaching a separate Python course.
How this course works
Intuition → math → code
Each idea starts in plain language, then gets precise with math, then becomes code.
See it work in Python
Read the explanation, then copy the example into your own Python environment. Change an input and explore what happens.
Watch, practise, reflect
Companion videos explain the same ideas visually. Practice and short quizzes help you check your understanding.
The learning journey
Planned topics, from foundations to applications. Lessons will become available as they are completed and reviewed.
- Module 1Coming soon
Orientation
The AI landscape, the vocabulary you need, and how machine learning works end to end.
- Module 2Coming soon
Math Essentials
Reading math notation from scratch, then just enough vectors, matrices, derivatives, exponentials, and probability to read ML math comfortably.
- Module 3Coming soon
Working with Data
NumPy and pandas, cleaning data, features and scaling, train/validation/test splits, and data leakage.
- Module 4Coming soon
Supervised Learning
Regression and classification with linear and logistic regression, k-nearest neighbours, decision trees, random forests, and SVMs.
- Module 5Coming soon
Evaluating and Improving Models
Metrics for regression and classification, cross-validation, overfitting, regularisation, and tuning.
- Module 6Coming soon
Unsupervised Learning
Finding structure without labels with clustering, dimensionality reduction (PCA), and anomaly detection.
- Module 7Coming soon
Neural Networks
Perceptrons, multilayer networks, backpropagation, and training loops in PyTorch.
- Module 8Coming soon
Convolutional Networks
Convolutions, pooling, image classifiers, and transfer learning.
- Module 9Coming soon
Sequences and the Road to Attention
Embeddings, recurrent networks, and the problem attention was invented to solve.
- Module 10Coming soon
Transformers in Depth
Self-attention, multi-head attention, the transformer block, and BERT vs GPT.
- Module 11Coming soon
Large Language Models
Pretraining, fine-tuning, alignment basics, prompting, embeddings, and retrieval-augmented generation (RAG).
- Module 12Coming soon
Vision Transformers
Treating images as sequences of patches, and how ViTs compare with CNNs.
- Module 13Coming soon
Object Detection
Bounding boxes, IoU, non-max suppression, mAP, and the YOLO, Faster R-CNN, and DETR families.
- Module 14Coming soon
Generative Models
Models that create new data, with autoencoders, GANs, and diffusion models.
- Module 15Coming soon
Reinforcement Learning
Learning from rewards, with agents, environments, and an introduction to Q-learning and policy methods.
- Module 16Coming soon
Responsible AI
Bias and fairness, privacy, safety, and how evaluations can mislead.
- Module 17Coming soon
Capstone
Put everything together in an end-to-end project.