ICT Architect (A)

Project overview

For our client, a renowned bank based in Zurich, we are looking for an experienced personality as Senior AI Security Architect. In this position, you will be part of an interdisciplinary team of security, architecture, engineering, and business departments and actively shape the secure use of modern AI technologies. The focus is on developing a holistic security architecture for LLM, RAG, and agentic AI applications, identifying and securing AI-specific risks, and designing Security-by-Design, Identity, Authorization, and Guardrail concepts.

This will allow you to make a central contribution to operating innovative AI use cases securely, controllably, and in compliance with regulations in a demanding banking environment.

Please note that due to current Swiss legislation, we can only consider applications from Swiss citizens, EU citizens, and individuals with a valid work permit.

Remote work from abroad is not possible.

Responsibilities

• Conception and further development of a holistic security architecture for AI use cases, especially for LLM, RAG, and agentic applications
• Analysis of new attack surfaces and trust boundaries between users, models, AI agents, tools, data sources, and external services
• Development of AI-specific threat models as well as preventive and detective security controls
• Evaluation and securing of risks such as prompt injection, jailbreaking, model manipulation, data poisoning, insecure output handling, data leakage, and security-relevant hallucinations
• Development of Security-by-Design concepts for RAG architectures, retrieval processes, vector databases, and connected knowledge sources
• Ensuring existing access and authorization specifications for AI-based data access and preventing cross-user and cross-tenant leakage
• Design of identity, authorization, and trust concepts for users, models, agents, and tools according to the principle of least privilege
• Conception of secure agent-to-tool and agent-to-agent communication as well as control of autonomous actions and critical transactions
• Development of reusable guardrails and AI-specific security controls, including for input/output validation, prompt and context protection, content filtering, and policy enforcement
• Definition of security mechanisms for tool/plugin approvals, secrets handling, sandboxing, and controlled use of agent capabilities
• Conception of suitable monitoring, logging, and governance mechanisms
• Translation of technical AI risks into comprehensible architecture, security, and control requirements
• Close collaboration with architecture, security, engineering, operations, business departments, and stakeholders from business and IT management

Requirements

Soft Skills

• Several years of professional experience as a Security Architect, ideally with a focus on Cloud Security and/or Application Security
• At least 5-8 years of relevant professional experience at a senior level
• Profound technical understanding of modern AI platforms, Large Language Models (LLMs), RAG architectures, and agentic workflows
• Demonstrable experience in threat modeling and in implementing Security-by-Design concepts
• Experience in developing and implementing guardrails, access and authorization concepts, as well as monitoring and logging approaches for complex applications
• Profound understanding of AI-specific security risks such as prompt injection, jailbreaking, data leakage, model manipulation, and security-relevant hallucinations
• Ability to translate technical risks into concrete architecture and control requirements and to accompany their implementation
• Experience in communicating complex security topics to technical and non-technical stakeholders
• Advantageous:
• Practical experience with securing productive LLM, RAG, or AI agent platforms
• Experience with AI applications in banks, insurance companies, or other highly regulated industries
• Knowledge of relevant regulatory and organizational requirements in AI Governance, Information Security, and Data Protection
• Experience with cloud-based AI and security architectures
• German at C2 level or near-native proficiency, as well as very good English skills
• Strong analytical and conceptual thinking skills
• High security and risk awareness
• Independent, structured, and solution-oriented way of working
• High degree of personal responsibility and commitment
• Very good communication and presentation skills
• Ability to convey complex technical issues understandably and appropriately for the audience
• Confident stakeholder management at both technical and management levels
• Teamwork skills and enjoyment of collaboration in interdisciplinary teams
• Ability to constructively reconcile different interests between security, engineering, architecture, and business
• Willingness to continuously engage with new technologies, attack scenarios, and security requirements in the dynamic AI environment

Project details

• Start: 2026-11-23
• Location: Zürich / Remote (nur aus der Schweiz), CH
• Languages: German