Just enough ML to be dangerous (in a good way)
Demystify language models without the heavy math
Common failure modes and how to spot them
Understanding the fundamental tension in AI development