AI Security Engineer (Red Teaming)

Company

hirify.global is a values-led, woman-owned delivery partner focused on empathy, integrity, and technical growth.

Location: Must be based in Canada

What you will do

• Perform hands-on adversarial testing including multi-turn jailbreaks, prompt injection, and API abuse.
• Analyze edge-case findings from AI red-team campaigns to reproduce and document vulnerabilities.
• Map security findings to industry frameworks like OWASP Top 10 for LLM, NIST AI RMF, and MITRE ATLAS.
• Provide actionable remediation guidance and conduct retesting to ensure security fixes are effective.
• Translate complex technical findings into clear risk assessments for both engineering and non-technical stakeholders.

Requirements

• Expert-level Python programming with deep proficiency in PyTorch, TensorFlow, and Hugging Face.
• Hands-on experience fine-tuning ML models and SLMs using techniques like LoRA/QLoRA and PEFT.
• Proven ability to design and execute adversarial attacks such as data poisoning and model extraction.
• Experience with adversarial ML toolkits like ART, CleverHans, and Foolbox.
• Strong threat-modeling skills and an attacker's mindset.
• Familiarity with MLOps practices including model versioning and secure deployment pipelines.

Culture & Benefits

• Fully remote work environment based in Canada.
• Clear career progression path toward staff and principal-level technical roles.
• Full support throughout the hiring process and ongoing professional accountability.
• Collaborative, values-driven team culture.