Adversarial Machine Learning Engineer – Red Teaming

This is a fully remote position, open to applicants in Canada.

📋 Description

• Perform hands-on adversarial testing across models, applications, agentic layers, and data pipelines, which includes multi-turn jailbreaks, guardrail bypasses, prompt injection, agent and tool-chain misuse, dangerous-capability assessment, API abuse, data poisoning, model inversion, and membership inference.

• Analyze edge-case findings from AI red-team initiatives and transform identified anomalies into recognized, reproducible vulnerabilities.

• Generate severity-ranked findings aligned with the OWASP Top 10 for LLM Applications, NIST AI Risk Management Framework and Generative AI Profile, MITRE ATLAS, and the EU AI Act Article 55 requirements.

• Provide evidence and detailed reproduction steps for discovered vulnerabilities.

• Offer practical remediation advice and carry out retesting to confirm fixes.

• Collaborate closely with the client’s Guardrails and AI red-teaming teams.

• Convey technical findings in accessible language for both technical engineers and non-technical stakeholders.

• Stay involved throughout the remediation process and final retesting.

⛳️ Requirements

• Advanced Python programming expertise with extensive knowledge of ML frameworks such as PyTorch, TensorFlow, and Hugging Face Transformers.

• Practical experience in fine-tuning ML models and Small Language Models (SLMs), which includes LoRA/QLoRA, PEFT, instruction tuning, and domain adaptation, aimed at enhancing performance and robustness.

• Strong understanding of ML mathematics, including optimization, linear algebra, probability, and statistics.

• Demonstrated ability to design and implement adversarial attacks, such as evasion, data poisoning, model extraction, and membership inference.

• Experience in applying defenses like adversarial training, robust fine-tuning, input sanitization, and differential privacy.

• Familiarity with adversarial ML toolkits like Adversarial Robustness Toolbox (ART), CleverHans, and Foolbox.

• Experience in red-teaming AI/LLM systems, encompassing prompt injection, jailbreak testing, and safety/alignment assessments.

• Capability to evaluate and benchmark model robustness, safety, and security posture both before and after fine-tuning.

• Knowledge of MLOps practices, including model versioning, experiment tracking, and secure deployment pipelines.

• Strong threat-modeling abilities and an attacker’s perspective, with the skill to communicate risks effectively to both technical and non-technical stakeholders.

• Current awareness of the latest developments in adversarial ML and GenAI security research.

🏝️ Benefits

• Fully remote work opportunity from anywhere in Canada, focused on delivery rather than mere presence.

• A clear trajectory for advancement into staff and principal-level technical roles.

• Comprehensive support from C-Serv throughout the hiring process and beyond, with complete accountability.

• A values-driven, woman-owned delivery partner founded on principles of empathy, integrity, collaboration, and growth.