Deep Learning & Modern AI

Mission. Neural networks from first principles through modern practice.

Scope

Learning objectives

Lessons

This track is scaffolded and waiting for content. See the curriculum dashboard for the roadmap and progress across all tracks.

Planned lessons (21)

Wave 3 — depth (provisional roadmap) — 21 lessons

foundations

  • The Perceptron concept · beginner · M A linear model with a learning rule, and the boundary of what it can learn.
  • Backpropagation by Hand concept · advanced · L The chain rule applied to a small network, on paper.
  • Backprop in NumPy example · advanced · L A tiny neural network with the backward pass written out.

pytorch

  • PyTorch Tensors and Autograd concept · intermediate · M The tensor, the graph, and the automatic gradient.
  • Training Loops in PyTorch concept · intermediate · M Batches, forward, backward, step — the loop you'll write a thousand times.

training

  • Overfitting and Dropout concept · intermediate · M The failure mode of deep nets and the regularization that fights it.
  • Optimizers: SGD to Adam concept · intermediate · M Momentum, adaptivity, and the defaults everyone uses.
  • Learning Rates and Schedules concept · intermediate · M The most important hyperparameter, tuned without superstition.

architectures

  • Convolutions concept · intermediate · M Locality and weight sharing: why CNNs see images.
  • CNN Architectures in Practice concept · intermediate · M Blocks, transfer, and not designing from scratch.
  • Embeddings concept · intermediate · M Meaning as geometry: entities as vectors.
  • Sequence Models: RNNs concept · intermediate · M Recurrent state, its limits, and why attention replaced it.
  • Attention concept · advanced · M Queries, keys, values: weighted lookup as a differentiable operation.
  • The Transformer concept · advanced · L The architecture behind the modern AI wave, layer by layer.

practice

  • Transfer Learning concept · intermediate · M Start from pretrained weights; fine-tune deliberately.
  • Fine-Tuning a Pretrained Model concept · intermediate · M Heads, freezing, and catastrophic forgetting on your own data.
  • Evaluating Neural Nets concept · intermediate · M Honest evaluation when the model has millions of parameters.

llms

  • LLM Application Patterns concept · intermediate · M Prompts, tools, agents: the patterns behind LLM-backed products.
  • Retrieval-Augmented Generation concept · intermediate · M Grounding LLM output in your documents, with evaluation.
  • Serving and Monitoring Models concept · advanced · M From checkpoint to endpoint, with feedback loops.

projects

  • Case Study: Image Classifier End to End project · intermediate · L Data, transfer learning, evaluation, and deployment.