Staff Security Engineer, Cloud & AI Platform

Brookfield Asset Management•Mount Tabor, VT

6d•Remote

About The Position

The Staff Security Engineer, Cloud & AI Platform is a deeply technical, hands-on role responsible for making Brookfield Real Estate’s cloud and AI platforms secure by design. You will lead security architecture and implementation across AWS, Infrastructure as Code, software delivery, internally developed applications, and AI/agent systems. This is a builder role. You will threat-model a design, review the underlying Terraform and application code, implement the control, instrument it, and help teams adopt it without unnecessary friction. Security requirements become reusable modules, automated checks, secure defaults, and clear engineering guidance rather than documents and findings lists. You will partner closely with Cloud Solutions, application engineering, AI engineering, Identity, CyberOps, and GRC/Privacy. Platform engineering continues to own general platform uptime, capacity, and deployment mechanics, and CyberOps continues to own alert triage; this role owns the architecture, controls, tooling, and assurance mechanisms that make those systems safe.

Requirements

Demonstrated senior or Staff-level ownership of production security or platform systems, including architecture decisions others build on
Deep AWS security experience across IAM, multi-account or AWS Organizations environments, logging and detection, KMS, networking, and service-to-service authorization
Strong Terraform or OpenTofu skills, including modular design, remote execution, policy controls, and safe state-aware changes
Experience securing CI/CD systems such as GitHub Actions, including OIDC federation, runner trust boundaries, secrets, artifacts, and deployment permissions
Working knowledge of application security fundamentals: threat modeling, authentication and authorization, secure API design, secrets handling, dependency risk, and vulnerability remediation
Ability to read and write production-quality automation or application code in Python, Go, Ruby, or a comparable language
Demonstrated ability to influence teams you do not manage, make pragmatic risk decisions, and turn ambiguous security needs into implemented controls

Nice To Haves

Experience with Amazon Bedrock, AgentCore, MCP, LLM gateways, AI guardrails, or production agentic systems
Container and orchestration security — ECS/Fargate or EKS, image supply chains, runtime isolation, and egress control
Identity platform depth: IAM Identity Center, Okta, Delinea, SAML, OIDC, OAuth, SCIM, ABAC, and enterprise entitlement systems
Experience with Scalr, Terraform Cloud, or another remote Infrastructure as Code platform
Familiarity with CrowdStrike Falcon Cloud Security, Datadox, CloudWatch, or comparable security and observability tooling
Data platform and sensitive-data controls involving S3, Snowflake, PostgreSQL/Aurora, or vendor data ingestion
AI threat modeling, red teaming, or security evaluation experience

Responsibilities

Help define security architecture for the cloud and AI platforms: set direction, write specs, and steward secure-by-default patterns across teams
Define security engineering standards and paved roads, document important decisions, and communicate risk and trade-offs to engineering and business leadership
Review infrastructure and application designs for authorization, network, data-protection, secrets, and tenant-isolation risks
Mentor engineers and raise the organization’s ability to make sound security decisions independently
Partner with Cloud Solutions, application engineering, AI engineering, Identity, CyberOps, and GRC/Privacy to prioritize work and drive adoption
Own and evolve security architecture for a multi-account AWS organization — account boundaries, service control policies, IAM Identity Center, delegated security services, KMS, VPC controls, and WAF
Design least-privilege human and workload access across AWS, Okta, GitHub Actions, and Infrastructure as Code platforms using federation and short-lived credentials
Build and maintain reusable security controls in Terraform or OpenTofu, with safe rollout, testing, monitoring, and recovery patterns
Strengthen centralized logging and evidence — CloudTrail, Config, GuardDuty, Security Hub, Macie, and CloudWatch — and keep guardrails current as the environment grows
Establish secure GitHub Actions and runner patterns, including OIDC trust, environment separation, workflow protection, action pinning, artifact integrity, and least-privilege deployment roles
Integrate practical security checks into developer workflows — secrets detection, dependency and container scanning, SBOMs, Infrastructure as Code analysis, code scanning, and risk-based release gates
Support vulnerability management for internally developed software, from detection and prioritization through remediation evidence and exception handling. Partner with the Vulnerability Manager to establish prioritization and reduce false positives
Improve controls with engineers rather than around them, avoiding noisy or easily bypassed gates
Lead threat modeling and security design reviews for web applications, APIs, data ingestion, authentication and authorization flows, and agentic systems
Define secure patterns for Amazon Bedrock and other model platforms, including model access, guardrail lifecycle, invocation logging, usage attribution, and sensitive-data controls
Harden agent tools, MCP servers, gateways, sandboxes, and code-execution environments against prompt injection, confused-deputy attacks, excessive agency, secret disclosure, data exfiltration, and cross-tenant access
Build repeatable security evaluations and adversarial tests for AI-enabled features and connect findings to engineering remediation
Translate cloud and application telemetry into actionable detections, runbooks, ownership, and escalation paths in partnership with CyberOps
Lead technical response and root-cause remediation for cloud, identity, software supply chain, and AI security incidents
Establish severity and ownership standards for cloud findings, software vulnerabilities, and AI security issues, with clear remediation expectations
Produce the control evidence GRC and Privacy need, without turning engineering work into manual reporting