Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

Video course · 15 chapters · 187 min · certificate

Become an AI Researcher: Math, PyTorch, LLMs & Transformers

A step-by-step path into AI research: functions, derivatives, vectors, gradients, matrices and probability; PyTorch tensors, reshaping and indexing; neural networks and backpropagation from scratch; and transformers: attention, causal self-attention, RoPE, multi-head attention, transformer blocks and tokenization.

Become a memberAlready a member? Sign in

What you'll learn

  • Master the core math for deep learning
  • Work with PyTorch tensors
  • Build neural networks and understand backpropagation
  • Explain attention with queries, keys and values
  • Implement multi-head and causal self-attention with RoPE
  • Assemble a transformer block and tokenize text

Chapters

15 chapters · 186:52
  1. 10:48 01Intro Members Welcome and setup Course overview, requirements and setup.
  2. 8:22 02Math Members Functions Linear, quadratic, cubic and square root functions.
  3. 14:09 03Math Members Derivatives Derivatives as rate of change.
  4. 12:48 04Math Members Vectors Magnitude, dot product and normalization.
  5. 8:56 05Math Members Gradients Steepest ascent/descent and partial derivatives.
  6. 13:36 06Math Members Matrices Multiplication, transpose and identity.
  7. 19:09 07Math/PyTorch Members Probability and tensors Expected value and conditional probability; creating tensors.
  8. 13:14 08PyTorch Members Squeezing and unsqueezing Adding and removing dimensions.
  9. 8:53 09PyTorch Members Indexing and slicing Selecting tensor elements.
  10. 13:12 10Neural nets Members Special tensors and neurons Zeros, ones, linspace, and coding a neuron.
  11. 11:07 11Neural nets Members Multi-layer networks and backprop Layers of neurons and backpropagation.
  12. 18:25 12Transformers Members Attention: queries, keys, values The attention mechanism.
  13. 11:28 13Transformers Members Self-attention, causal masks and RoPE Self-attention, causal self-attention and rotary embeddings.
  14. 10:56 14Transformers Members Multi-head attention Splitting into multiple heads.
  15. 11:49 15Transformers Members Transformer block and tokenization Feed-forward, add & norm, tokenization and conclusion.