Tensor by TensorA clear path from foundations to systems
Learn AI in layers.
Start with the small picture. Open the detailed notes when you are ready to see how the pieces fit.
Start learning →No account. Your progress stays on this device.
0modules
0lessons
0%complete
THE LEARNING PATH
Choose a module and follow the nodes.
Each lesson has a quick ELI5 view and a deeper reading path.
WORK IT OUT, THEN BUILD IT
Deeper studies, labs, and papers.
Follow the curriculum first, then test your understanding with a complete worked study.
- Why softmax cross-entropy produces p − y
Foundation · Derive the gradient, calculate a small example, and check it with autograd.
- Build causal attention with a correct incremental KV cache
Hands-on lab · Test masks, cache positions, and full-sequence equivalence.
- Attention Is All You Need: a careful paper walkthrough
Research reading · Connect the original Transformer architecture to a small runnable experiment.