Career-Aware Resume Tailoring via Multi-Source Retrieval-Augmented Generation with Provenance Tracking: A Case Study

TL;DR

This paper introduces Resume Tailor, a system that uses multi-source retrieval-augmented generation with provenance tracking to improve resume tailoring by leveraging a longitudinal career database, showing promising results in specific job categories.

Contribution

The paper presents a novel resume tailoring system that integrates multi-source retrieval-augmented generation with provenance tracking and longitudinal career data for more grounded and personalized suggestions.

Findings

Retrieval-augmented generation improved fit scores by 7.8 points for relevant roles.

Scores decreased by 8.0 points when domain-specific experience was absent.

Longitudinal retrieval benefits resume tailoring when prior relevant experience exists.

Abstract

AI-assisted resume tailoring systems commonly operate on a single uploaded resume, which limits their ability to recover relevant experience omitted from the current draft and makes it difficult for users to distinguish grounded edits from model-generated suggestions. This paper presents Resume Tailor, an agentic resume-tailoring system that maintains a longitudinal career vault in a vector database and uses multi-source retrieval-augmented generation (RAG) to assemble job-specific resume content from historical resumes and structured career records. The system is implemented as a 12-node LangGraph pipeline with typed state management, hybrid semantic-lexical confidence scoring, provenance-aware fallback generation, anti-hallucination guardrails, and a conditional review loop. We report a pilot evaluation on nine job descriptions (JDs) across software engineering, data analytics, and business analysis roles using a single candidate's career history. For six JDs where the candidate held at least one prior role in the same occupational category, enabling the career vault improved Applicant Tracking System (ATS)-style fit scores by an average of 7.8 points. For two JDs requiring domain-specific expertise absent from the vault, scores decreased by an average of 8.0 points. One partially overlapping role showed a modest gain of 2 points. These results suggest that longitudinal retrieval can improve resume tailoring when relevant prior experience exists, while also highlighting the need for confidence-gated retrieval when domain overlap is weak.