import re def clean_text(text): """Basic text cleaning""" text = text.lower() text = re.sub(r'[^\w\s]', '', text) # Remove punctuation text = re.sub(r'\s+', ' ', text) # Remove extra whitespace return text.strip() def extract_skills(text): """Extract skills from text (simplified version)""" common_skills = { # Technical skills 'python', 'java', 'javascript', 'sql', 'html', 'css', 'react', 'docker', 'kubernetes', 'aws', 'machine learning', 'tensorflow', 'pytorch', 'git', 'linux', 'node.js', 'typescript', 'azure', 'gcp', # Soft skills 'communication', 'leadership', 'problem solving', 'teamwork', 'project management', 'agile', 'scrum' } found_skills = set() for skill in common_skills: if skill in text: found_skills.add(skill.title()) return list(found_skills)