Senior AI Security Engineer

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

📋 Description

• Design and establish guardrails for AI systems that utilize tools, including:

• Access controls and allowlists for tools

• Isolation of context and memory

• Validation of agent actions at each step

• Implement mitigations that align with the OWASP Agentic AI Top 10 (e.g., prompt injection, unsafe tool usage, data leakage, excessive autonomy)

• Create enforcement mechanisms that regulate AI behavior during execution:

• Interceptors, proxies, or middleware for tool/API interactions

• Layers for policy decision and enforcement

• Rate limits, execution bounds, and emergency shutdowns

• Design and enforce identity and access controls for agents and automation, which include:

• Temporary credentials and scoped permissions

• Distinct separation between human and non-human access

• Strong linkage of identity to task context and execution

• Ensure that all AI actions can be traced and attributed

• Implement logging and tracing mechanisms for AI activities:

• Prompts, tool usage, and decision processes

• Develop detection capabilities through:

• Behavioral baselining and anomaly detection techniques

• Identify and notify on:

• Unusual tool usage

• Questionable prompt patterns

• Unexpected data access

• Conduct threat modeling for agentic systems using MAESTRO, which involves:

• Mapping agent capabilities, trust boundaries, and potential attack routes

• Modeling misuse and adversarial situations

• Converting findings into actionable safeguards and detection mechanisms

• Safeguard developers utilizing AI tools by:

• Preventing exposure of sensitive data

• Validating AI-generated code and actions

• Restricting unsafe automation

⛳️ Requirements

• Over 7 years of experience in security engineering or backend systems

• Demonstrated experience in designing and implementing security controls, such as:

• Runtime enforcement layers (proxies, middleware, policy engines)

• Identity and access management systems, particularly for non-human entities

• Excellent programming abilities (Python preferred; familiarity with Go, Java, or TypeScript is a plus)

• Experience with AI-assisted development tools like Claude Code in real-world applications, including an understanding of related security risks and safeguards

• Experience with:

• Logging, monitoring, and detection systems

• Creating or securing API/service interactions

• Practical knowledge of:

• Agentic AI systems or tool-integrated LLM workflows

• OWASP guidance related to AI/agent risks.

🏝️ Benefits

• Health insurance

• 401(k) matching

• Flexible work hours

• Paid time off

• Remote work options