Hiring workflow automation

CV Recruiter RAG Agent

Status: Research / prototype
Role: Engineer
Organisation: Independent
Year: 2025

The problem

Screening a stack of CVs against one job description is slow, inconsistent, and the reasoning behind a rejection is never written down.

Outcome

Scoring: Citation-backedScoringEach ranking traces to source text.
Inputs: PDF + DOCXInputs

The CV Recruiter ranking view: a job description on the left, candidates ranked by match score on the right, each score citing the exact line in the CV that earned it.

Approach

01
Parses PDF and DOCX CVs into a vector store, then ranks candidates against a job description.
02
Every score is citation-backed - the agent points at the line in the CV that justified it, so a decision can be audited.
03
Orchestrated with LangGraph over the Claude API, served through FastAPI.

Stack

LangGraph
Claude API
FAISS
FastAPI
Python