How This AI Works
This is a Retrieval-Augmented Generation,
or RAG, application.
1
You ask a question
Django receives and validates your request.
2
Relevant information is found
OpenAI embeddings, PostgreSQL, and pgvector
identify the strongest matching career content.
3
A grounded prompt is created
Retrieved résumé, project, and testimonial
content is combined with accuracy instructions.
4
The answer is generated
OpenAI writes a professional response using
the retrieved career evidence.
Why use RAG?
It helps prevent unsupported answers by grounding
each response in Ryan’s actual career documents.