Two depths of understanding

Follow the broad path for orientation. Follow the systems path until you can predict, measure, and defend a deployment decision. Progress by evidence of understanding, not a fixed number of weeks.

Understand the AI landscape

Recognize the major ideas, explain their mechanism in ordinary language, and know where to go deeper. Read the mental picture, explore the selected part of a reference, then answer the check in your own words.

7 modules · 32 study maps

  1. 01 · The AI landscape
  2. 02 · Just enough mathematical intuition
  3. 03 · How learning succeeds and fails
  4. 04 · Inside a neural network
  5. 05 · Beyond text
  6. 06 · How LLMs work
  7. 07 · Acting, retrieving, and knowing the limits

Finish line: You can explain how search, learning, perception, generation, and action fit together without deriving every equation.

Run LLMs at scale

Follow a request from the gateway to GPU memory and back. Predict a bottleneck, study the mechanism, run an experiment, and defend the tradeoff. Begin after you can explain tokens, attention, and the generation loop.

10 modules · 53 study maps

  1. 08 · One request, end to end
  2. 09 · Why the GPU waits
  3. 10 · Keep useful work on the GPU
  4. 11 · Spend fewer bytes and steps
  5. 12 · Beyond one GPU
  6. 13 · Operate a serving fleet
  7. 14 · Serve whole AI applications
  8. 15 · Training systems you need to recognize
  9. 16 · Measure, protect, and pay for the service
  10. 17 · Build evidence of engineering judgment

Finish line: You can size, benchmark, optimize, deploy, and troubleshoot a serving system against explicit quality, latency, reliability, and cost targets.