DEV Community

The Answerability Problem Series' Articles

Back to Giulio D'Erme's Series
The AI-memory benchmark everyone quotes forbids saying “I don't know”
Cover image for The AI-memory benchmark everyone quotes forbids saying “I don't know”

The AI-memory benchmark everyone quotes forbids saying “I don't know”

2
Comments 2
7 min read
Relevance is not answerability: six signals, and none of them beat plain cosine
Cover image for Relevance is not answerability: six signals, and none of them beat plain cosine

Relevance is not answerability: six signals, and none of them beat plain cosine

Comments 2
9 min read
“Does your agent know what it doesn’t know?” has no answer. It has a coordinate.
Cover image for “Does your agent know what it doesn’t know?” has no answer. It has a coordinate.

“Does your agent know what it doesn’t know?” has no answer. It has a coordinate.

1
Comments 8
7 min read
The memory layer that never calls an LLM: what that buys, and what it costs
Cover image for The memory layer that never calls an LLM: what that buys, and what it costs

Raw turns beat LLM cost via auditability

The memory layer that never calls an LLM: what that buys, and what it costs

20
Comments 30
10 min read
The RAG benchmark that pays you for saying "I don't know", and what it exposed
Cover image for The RAG benchmark that pays you for saying "I don't know", and what it exposed

The RAG benchmark that pays you for saying "I don't know", and what it exposed

Comments
7 min read