Full job description

ARE YOU READY to step into the New Era (NewRA) of AI-driven banking?

At Accenture Newra AI Hub, we are not just building technology, we are redefining how banking operates. As part of our strategic collaboration with Piraeus, Newra is designed to responsibly embed AI at the core of its business, moving beyond experimentation to real-world impact at scale. Built to make a real difference, Newra reflects our belief that AI creates value only when it genuinely improves people’s lives.

You will work on advanced AI solutions that span the full spectrum of the bank -from core banking systems to customer experience- simplifying complexity, automating critical processes and delivering measurable results where they matter most.

Joining Newra means becoming part of a high-performing team of innovators at the beginning of a major reinvention. This is a space for people who approach AI with depth, discipline and purpose. You will collaborate across disciplines, develop future-proof skills, and help turn technology into real-world transformation.

As an AI Security Engineer, y ou’ll help secure enterprise AI and GenAI systems from design through production, making sure models, prompts, agents, data flows, and integrations are protected against misuse, leakage, manipulation, and abuse. You’ll work with product, engineering, security, risk, compliance, and Responsible AI teams to threat-model AI use cases, define technical guardrails, coordinate security testing and red teaming, and embed secure-by-design controls into the AI delivery lifecycle.

What you'll build

AI security patterns and control requirements that help teams design, build, deploy, and monitor AI systems securely and consistently
Threat models for GenAI, agentic AI, machine learning, RAG, model endpoints, data pipelines, and third-party AI platforms
Guardrail and runtime-defense specifications embedded into AI products — input/output filtering, prompt-injection detection, data-leakage prevention, jailbreak resistance, and least-privilege boundaries for agent tool use
AI security testing approach covering adversarial prompts, abuse cases, data exposure, insecure integrations, model/tool misuse, and pre-deployment security gates
Secure architecture checkpoints embedded into product, data, engineering, cloud, security, compliance, and release processes
Monitoring and response mechanisms for AI security events, including suspicious prompts, policy violations, sensitive-data exposure, tool misuse, and anomalous model behavior