Adversarial Machine Learning Engineer - Red Teaming
New
C
C-ServAI Security
Fully remote working anywhere in the CanadaFull-TimeMiddle
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Required Skills
- PythonPyTorchTensorflowLLMMLOps
Requirements
- Expert-level Python programming proficiency.
- Deep proficiency in ML frameworks such as PyTorch, TensorFlow, and Hugging Face Transformers.
- Hands-on experience fine-tuning ML models and Small Language Models (SLMs) using PEFT, LoRA/QLoRA, and instruction tuning.
- Strong foundation in ML mathematics, including optimization, linear algebra, probability, and statistics.
- Proven ability to design and execute adversarial attacks like evasion, data poisoning, model extraction, and membership inference.
- Experience implementing defensive measures like adversarial training, robust fine-tuning, and differential privacy.
- Proficiency with adversarial ML toolkits such as ART, CleverHans, and Foolbox.
- Experience red-teaming AI/LLM systems, specifically prompt injection, jailbreak testing, and safety evaluation.
- Familiarity with MLOps practices, including model versioning and secure deployment pipelines.
- Strong threat-modeling skills with an attacker's mindset.
Responsibilities
- Conduct hands-on adversarial testing across models, applications, the agentic layer, and data pipelines.
- Investigate edge-case findings from red-team campaigns to turn anomalies into reproducible vulnerabilities.
- Map findings to industry frameworks including OWASP Top 10 for LLM, NIST AI RMF, MITRE ATLAS, and EU AI Act.
- Document findings with evidence and clear reproduction steps.
- Provide actionable remediation guidance and perform retests to confirm fixes.
- Work embedded with client teams to communicate technical risks to both technical and non-technical stakeholders.
View Full Description & ApplyYou'll be redirected to the employer's site
