Visual Overview

Barba-CV provides a deterministic structured layer between unstructured CV documents and AI / ATS systems.

The Problem

CVs are unstructured documents with highly variable layouts and content styles, which makes consistent machine processing difficult.

Common issues include:

  • PDF parsing problems
  • inconsistent formatting
  • ATS incompatibilities
  • difficulty for AI systems to reliably extract structured data

The Barba-CV Approach

Barba-CV introduces:

  • a deterministic JSON structure
  • a schema describing the structure
  • a flexible semantic layer for real-world CV variability

Conceptual Pipeline

CV document
↓
Text extraction
↓
LLM mapping using Barba-CV example JSON
↓
Barba-CV structured JSON
↓
HR systems / ATS / AI agents / MCP services

Why this matters for AI systems

Barba-CV enables:

  • deterministic structured CV data
  • interoperability between AI systems
  • reliable HR data pipelines
  • schema-driven parsing