AI Agentic Security Engineer

Staff AI Agentic Security Engineer

New York City

Bridgewater Associates is a premier asset management firm, focused on delivering unique insight and partnership for the most sophisticated global institutional investors.

Our investment process is driven by a tireless pursuit to understand how the world's markets and economies work — using cutting edge technology to validate and execute on timeless and universal investment principles.

Founded in 1975, we are a community of independent thinkers who share a commitment for excellence. By fostering a culture of openness, transparency, and inclusion, we strive to unlock the most complex questions in investment strategy, management, and corporate culture.

The Security Department's mission is to protect Bridgewater. We constantly evolve our cyber, physical, and staff security practices to meet business needs and stay ahead of the changing threat landscape.

This person needs to know how to build and how to protect. We're not looking for someone who reviews architectures from the sideline. We need someone in the arena — writing agents, shipping code, deploying guardrails, and setting the standard for how an entire firm adopts AI securely.

This is a 50/50 role with two equally critical mandates:

PILLAR 1 (50%) — AI Thought Leader in Security:
Build It

• Build Security Operations Agents: Design, develop, and deploy autonomous agents for threat detection, alert triage, vulnerability management, and incident response — to transform the way those teams operate.
• Modernize Workflows AI-Natively: Reimagine existing security processes through the lens of agentic AI. Replace manual runbooks with intelligent agents that reason, act, and escalate. Build agent-powered security copilots for engineering teams that perform real-time code review, suggest secure patterns, and catch vulnerabilities before they ship.
• Own the Security AI Stack: Evaluate, select, and implement the right mix of frameworks, orchestration tools, and infrastructure for the department's agent platform. You should have strong opinions — backed by hands-on experience — on Lang Graph, Lang Chain, CrewAI, Auto Gen, OpenAI Agents SDK, Google ADK, Semantic Kernel, Dify, n8n, and the broader ecosystem.
• Governance and framework Automation: Build agents that continuously validate configurations, access policies, and data handling against regulatory and internal frameworks of the agents deployed by our investment teams.
• Be the agentic security thought leader: Be the person the department looks to for what's possible. Stay deeply current on the AI landscape — enterprise and open-source — and translate that knowledge into real capability.

PILLAR 2 (50%) — Forward-Deployed AI Security Architect:
Protect It

• Deep Architecture & Sandboxing: Design secure deployment architectures for AI agents across the firm. Define sandboxing strategies, execution boundaries, network isolation, and blast-radius controls that let teams move fast without exposing the organization to unacceptable risk.
• Identity & Authorization for Agents: Architect identity strategies for a world where agents act on behalf of humans. Define how agents authenticate, what permissions they hold, how credentials are scoped and rotated, and how to enforce least-privilege across multi-agent systems and MCP server integrations.
• AI Supply Chain Security: Own the security posture of the AI supply chain end to end. Evaluate the security of agent frameworks, MCP servers, skills/plugins, model providers, embedding pipelines, vector databases, and every dependency in between. Understand the attack surface of tools like Lang Graph, Lang Flow, Dify, n8n, Open Interpreter, Claude Code, Cursor, and similar agentic development environments.
• Prompt Injection & Model Manipulation Defense: Be the firm's leading expert on prompt injection, jail breaking, data poisoning, indirect injection via tool outputs, and agent manipulation attacks. Design and deploy runtime defenses using tools like NeMo Guardrails, Llama Firewall, LLM Guard, Open Guardrails, Guardrails AI, and custom detection layers.
• Runtime Safety & Governance: Build monitoring, kill switches, escalation triggers, and anomaly…