Platform Security Engineer (AI Agents)

About the role

Brain Co. builds AI-native operating systems for large, regulated institutions, using agents that run real workflows on top of its proprietary Atlas platform. It is hiring a Platform Security Engineer to build the guardrails that keep AI agents safe on real customer data. This is a builder role, not a policy or compliance role: you will design and ship the code that constrains what an agent can access, do and expose, working as a software engineer embedded with product and agent-platform teams.

Key responsibilities

• Build guardrail services that mediate agent actions, scoped permissions, tool-call validation and hard limits on what an agent can read, write or send
• Write production code for data access controls that keep customer PII and sensitive records within approved boundaries
• Build reusable guardrail libraries and SDKs so product engineers can add data protection and permissioning to new agent workflows
• Design detection and containment for agent failures, prompt injection, tool misuse and data exfiltration, including automated tests and red-team harnesses
• Instrument agents with tamper-evident audit logs and decision trails
• Review new agent capabilities with product, platform and ML engineering teams and flag missing guardrails
• Own the developer experience for guardrails: clear APIs, documentation and low-friction integration
• Help define and measure guardrail effectiveness, coverage and false positive/negative rates

Required profile

• 5 to 8 years as a software engineer shipping production systems, with meaningful time on security, data protection or trust and safety problems
• Experience with or strong knowledge of AI agents in production, tool use, function calling and orchestration frameworks such as LangChain or LangGraph
• Solid grasp of data protection fundamentals: PII handling, access control, encryption and least privilege
• Comfortable designing systems used by other engineers, with clear interfaces, sensible defaults and predictable failure modes
• Strong written English, able to write documentation and runbooks
• Nice to have: direct experience building guardrails for LLM/agent systems, background in regulated industries, familiarity with policy-as-code, secrets management or supply-chain security tooling, prior startup experience

Required skills

• General-purpose programming in Python, Go, TypeScript or similar, designing services and APIs others depend on
• Cloud experience (AWS, GCP or Azure) sufficient to build and deploy services securely
• Full SDLC awareness, designing guardrails that do not create friction for developers

What we offer

• Annual base salary range of $190,000 - $325,000, depending on skills, scope and experience
• Meaningful equity
• Country-specific benefits

To apply, send your CV and a brief note about a guardrail, safety control or security tooling you have built to the hiring contact.