NOTES: AI Engineering by Chip Huyen: Preface
Preface
- In 2012, the AlexNet authors noted:
“All of our experiments suggest that our results can be improved simply by waiting for faster GPUs and bigger datasets to become available.”
- Papers about LLMs early as the 1950s.
- Retrieval-augmented generation (RAG) built upon retrieval technology that powers search and recommender systems.
- She also wrote Designing Machine Learning Systems which has other insights that will prove valuable.
- This book is not a tutorial.
- The book is laid out to follow the lifecycle of building an AI application.
- Model evaluation is the hardest part of AI Engineering.
- The last chapter will bring everything together to cover building an AI application end-to-end.
Three key model response levers:
- “The instructions for how the model should behave”
- “The context the model can use to respond to the query”
- “The model itself”
Terms
- large language models (LLMs)
- large multimodal models (LMMs)