Question 1

Why do we use RAG instead of fine-tuning?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 2

Explain a full RAG pipeline step-by-step

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 3

What is chunking? Why is it important?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 4

What is similarity search?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 5

What is a vector database?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 6

Which embedding model did you use?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 7

What is prompt engineering?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 8

How do you expose your RAG model as an API?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 9

How do you make RAG faster?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 10

What is an AI agent?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 11

How is an agent different from a normal LLM call?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 12

What are tools in an agent?

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.

Question 13

Questions on your projects and task assigned

Accepted Answer

This question was asked at Trestle labs during a OA interview for the AI Engineer  role.