AI Security Architect

SoftServe is seeking an expert in security architecture focused on AI and cloud security, working with complex multi-cloud environments including AWS and GCP.

• 7+ years in security architecture or cloud security, designing secure multi-cloud environments
• Hands-on experience securing AI/ML and LLM-based systems at data, model, and infrastructure layers
• Strong knowledge of AI-specific threats such as prompt injection, data leakage, model abuse, and agent/tool misuse
• Understanding of AI infrastructure components including model hosting, vector databases, orchestration frameworks, and AI gateways
• Experience conducting security assessments, adversarial testing, threat modeling, or red-team style AI evaluations
• Familiarity with AI security frameworks and standards including NIST AI RMF, OWASP Top 10 for LLM Applications, MITRE ATLAS, and ISO/IEC 42001
• Experience with threat modeling and designing security controls, guardrails, and isolation for complex workloads
• Solid cloud security background in GCP and/or AWS, including IAM, network isolation, containers, and logging
• Experience defining security telemetry, detection use cases, and integrating logs into SIEM or real-time analytics platforms
• Ability to translate strategy into actionable engineering requirements
• Strong communication and stakeholder management skills; fluent English

Nice to have:

• Experience with Python
• Knowledge of ISO standards
• Stakeholder management skills

• Build the foundations of the AI-based security program, including roadmap, governance, and approval processes
• Define how AI can be safely used for security testing, including AI-assisted code scanning and vulnerability discovery integration
• Define AI security architecture and adapt suitable AI security frameworks
• Design threat models for AI workloads and define security controls for AI environments
• Design reference architectures for secure AI platforms deployed across AWS and GCP
• Define secure operating principles, deployment standards, and isolation requirements for frontier models
• Design guardrails and monitoring/logging requirements for AI systems
• Translate architecture and standards into implementation backlog and support delivery