Series
Sequenced reading paths. Each series builds on the last so concepts land in the right order.
- Series5 parts · ~59 min total
LLMs from the foundations
Start with how neural networks work, then walk through the history of language models, the transformer, and how LLMs became useful and turned into the platform we use today.
- 01How Neural Networks Work
- 02LLMs Explained, Part 1: How We Got Here
- 03LLMs Explained, Part 2: How the Transformer Works
- 04LLMs Explained, Part 3: How LLMs Got Useful
- 05LLMs Explained, Part 4: The Platform Era
Open series → - Series3 parts
GPUs and the hardware behind AI
Understand the chips that run modern AI. Start with why AI ended up on GPUs, then open one up to see SMs, threads, and tensor cores, and finish with the H100/H200/B200 family, NVLink, NVL72 racks, and the alternatives to NVIDIA.
- 01GPUs Explained, Part 1: How AI Found Its Hardware
- 02GPUs Explained, Part 2: Inside a GPU
- 03GPUs Explained, Part 3: The Hardware Behind Modern AI
Open series →