AI Security Engineer (DevSecOps for agent systems)

The company is looking for an AI Security Engineer to design and implement security for agent systems. Remote work. Salary discussed after interview.

What you'll do

• Design security architecture for agent systems: sandboxes for tool execution, circuit isolation, limits, kill switch
• Protection against prompt injection, data exfiltration, jailbreak at the architecture, filter, and output validation level
• Red team exercises against the company's own agents
• Review agent pipelines from the LLM engineering team to production deployment
• Control what is sent to external LLM providers

What we need

• 3-5 years in DevSecOps / AppSec / Security Engineering
• Strong Python, experience with Docker, Linux, CI/CD
• Practical experience with Threat modeling
• Secrets management (Vault or similar), IAM, sandbox environments
• Understanding of how LLMs and agent systems work: LangChain / LangGraph / MCP / Claude Code
• Willingness to design and build agents yourself, not just review others' code

What's in it for you

• Remote / hybrid
• Real production tasks, not paper security
• Direct access to top LLMs (Claude, GPT, open-source) and budgets for experiments
• A team where AI agents are a daily working tool, not a presentation for investors

Stack & operators