"""
Atlas-G Protocol - Profile Extractor
Parses resume content into structured professional profile data.
"""
import re
from typing import Dict, List, Any
def extract_professional_profile(resume_content: str) -> Dict[str, Any]:
"""
Extract structured professional profile from resume content.
Args:
resume_content: Raw resume text
Returns:
Structured profile object with summary, experience, and skills
"""
if not resume_content:
return {"error": "No content provided"}
return {
"summary": _extract_summary(resume_content),
"experience": _extract_experience(resume_content),
"skills": _extract_skills_categorized(resume_content)
}
def _extract_summary(content: str) -> str:
"""Extract professional summary section."""
match = re.search(r'PROFESSIONAL SUMMARY\s*\n(.*?)(?=\n\n|\nCORE EXPERTISE)', content, re.DOTALL)
if match:
return match.group(1).strip()
return ""
def _extract_experience(content: str) -> List[Dict[str, str]]:
"""
Extract experience history.
Note: This uses regex heuristics based on the specific resume format.
"""
experiences = []
# Find the Professional Experience section
exp_section_match = re.search(
r'PROFESSIONAL EXPERIENCE\s*\n(.*?)(?=\nPROJECT PORTFOLIO)',
content,
re.DOTALL
)
if not exp_section_match:
return []
exp_text = exp_section_match.group(1)
# Split by role blocks (heuristic: ALL CAPS lines followed by parens or content)
# Looking for patterns like "NEURAFLASH (Senior Consultant...)"
# or "INDEPENDENT AI ENGINEER (Current)"
# Simple line-based parser for robustness
lines = exp_text.split('\n')
current_role = None
buffer = []
for line in lines:
line = line.strip()
if not line:
continue
# Detect Header: "COMPANY (Role)" or "ROLE (Context)"
# Heuristic: Upper case start, contains parens
if re.match(r'^[A-Z][A-Z\s]+', line) and '(' in line and ')' in line:
# Save previous
if current_role:
current_role["description"] = "\n".join(buffer).strip()
experiences.append(current_role)
buffer = []
# Start new
parts = line.split('(', 1)
title_company = parts[0].strip()
details = parts[1].rstrip(')')
current_role = {
"title": title_company, # Mixed field in this resume format
"context": details,
"description": ""
}
elif current_role:
buffer.append(line)
# Save last
if current_role:
current_role["description"] = "\n".join(buffer).strip()
experiences.append(current_role)
return experiences
def _extract_skills_categorized(content: str) -> Dict[str, List[str]]:
"""Extract skills sections."""
skills = {}
# Extract CORE EXPERTISE
core_match = re.search(r'CORE EXPERTISE\s*\n(.*?)(?=\n\n|\nPROFESSIONAL)', content, re.DOTALL)
if core_match:
core_lines = [line.strip().lstrip('- ') for line in core_match.group(1).split('\n') if line.strip()]
skills["core"] = core_lines
return skills
