Instrutor de AI Security | Foco em Segurança de Machine Learning

Requisitos

• 2+ years of information security experience combined with hands-on machine learning or adversarial ML work
• Real-world experience in at least one of: red teaming/pentest of ML models (poisoning, evasion, extraction, inference), adversarial harde..
• Practical experience with scientific Python for ML (scikit-learn, PyTorch, pandas/numpy) and ability to train, evaluate and serve predict..
• Hands-on familiarity with ART (Adversarial Robustness Toolbox) and/or Foolbox to generate, evaluate and mitigate adversarial attacks
• Understanding of the ML pipeline (collection, training, validation, deploy, serving) as an attack surface, distinguishing predictive/disc..

Desejáveis

• Published research or conference talks on adversarial ML or ML security (NeurIPS workshops, USENIX Security, Black Hat, DEF CON AI Villag..
• Real experience protecting production ML systems for credit scoring, fraud detection or malware classification
• Knowledge of commercial MLaaS platforms (AWS SageMaker, Google Vertex AI, Azure ML, BigML) and their API exposure risks
• Creation of public technical content: articles, repositories, courses or talks about machine learning security
• Ability to turn real projects and professional experience into practical, accessible lessons