Job Search

We are seeking an AI Security Engineer (Agentic Systems) to help design, evaluate, and secure intelligent systems that can reason, plan, use tools, and operate with varying degrees of autonomy. In this role, you will work across security engineering, machine learning, and software development to identify emerging risks in agentic architectures and build practical safeguards. You will contribute to secure-by-design principles throughout the lifecycle, from research and prototyping through deployment, monitoring, and continuous improvement.

Your responsibilities will include threat modelling AI agents, tool-use workflows, orchestration layers, memory systems, data pipelines, and model integrations. You will develop and execute adversarial tests covering prompt injection, data exfiltration, insecure tool access, privilege escalation, identity abuse, supply-chain threats, model manipulation, and unsafe autonomous actions. You will create security controls such as permission boundaries, sandboxing strategies, policy enforcement, input and output validation, audit logging, human-approval checkpoints, and anomaly detection. You will also investigate incidents, communicate findings clearly, and collaborate with engineering, research, product, privacy, and risk teams to translate complex vulnerabilities into measurable remediation plans.

The successful candidate will have professional experience in application, cloud, product, or infrastructure security, together with a strong understanding of modern AI or machine-learning systems. Experience securing LLM applications, autonomous agents, APIs, distributed systems, or developer platforms is highly desirable. You should be comfortable programming in Python or a similar language and familiar with cloud security, identity and access management, secure software development, vulnerability assessment, and security testing frameworks. Knowledge of AI safety, model evaluation, red teaming, privacy engineering, or governance standards would be an advantage. We value curiosity, sound judgement, strong written and verbal communication, and the ability to work effectively in ambiguous, fast-moving environments. Candidates must be able to balance robust security with usability, reliability, and responsible innovation.