AI Product Engineering
Block 2: AI-Assisted Engineering & Integration
Model knows nothing after its training date
Model knows nothing about your domain, your documents, your data
Grounding = feed relevant information at query time, not from training data.
A list of numbers that represents the meaning of text.
Key property: similar meaning → similar numbers.
Real models: hundreds to thousands of dimensions.
| Score | Meaning |
|---|---|
| 1.0 | Identical meaning |
| 0.7–0.9 | Very similar |
| 0.5–0.7 | Related but different |
| 0.0 | Completely unrelated |
Find relevant documents = find highest similarity to query embedding.
Read each document → call embedding API → store text + vector
Embed query → find most similar vectors → feed matching text to LLM
Next Session: Building the RAG Pipeline