AI Software Security Engineer / Architect

BrightBox Group; United Kingdom; On-site; GBP 90,000 - 140,000; Full time

Job summary

BrightBox Group in the United Kingdom is seeking an AI Software Security Engineer/Architect to secure large-scale AI deployments. You will work across client programmes to design and implement robust security for LLMs and agentic systems, navigating legacy architectures to enable safe, compliant AI applications.

Responsibilities include building input/output controls, threat assessments, and governance aligned with ISO 27001 and NIST AI RMF, plus collaboration with cloud providers to secure

Qualifications

• Deep technical security expertise in cloud and application security.
• Proven experience deploying AI systems at enterprise scale.
• Strong Python coding skills and ability to learn additional languages.
• Ability to engage at executive levels and drive transformation programs.

Responsibilities

• Develop input/output controls and human-in-the-loop systems to secure LLM and agentic apps.
• Craft identity models for non-human actors to ensure safety and accountability.
• Conduct threat assessments focusing on data exfiltration and prompt injection.
• Align deployments with data protection frameworks like NIST AI RMF and ISO 27001.
• Design secure AI workloads across AWS, Azure, and Google Cloud.
• Support pre-sales initiatives and client presentations.
• Lead strategic security initiatives and governance for clients.

Skills

Python

Cloud security

AI security

Security architecture

Tools

AWS

Azure

GCP

NIST RMF

Job description

As an AI Software Security Engineer / Architect, you will become an integral part of our cyber practise. You will be deployed across client programmes where AI is unfolding and security structures are not yet mature. Your expertise will enable secure deployment of LLMs and agentic systems within large enterprises, navigating legacy architectures to establish robust security solutions.

Key Responsibilities:

Secure AI Design:

• Develop input/output controls, and human-in-the-loop systems to secure LLM and agentic applications.
• Craft identity models for non-human actors to ensure safety and accountability in AI interactions.
• Conduct thorough threat assessments on AI systems, focusing on vulnerabilities such as data exfiltration and prompt injection.

Risk and Governance:

• Align AI deployments with data protection regulations and frameworks like NIST AI RMF and ISO 27001.
• Establish governance for AI deployments, creating control catalogues and assurance processes.

Cloud and Alliance Integration:

• Design secure AI workloads across platforms such as AWS, Azure, and Google, ensuring data residency and key management.
• Collaborate with AI model providers to optimise enterprise deployment patterns.

Build and Prototype:

• Engage in the development of full-stack applications, automations, and agentic workflows within client settings.

Strategy and Advisory:

• Develop security strategies and operating models for client executives, converting technical insights into strategic propositions.
• Support pre-sales initiatives and client presentations.
• Deep technical security expertise in cloud security, application security, or related fields.
• Proven experience in AI system deployment and enhancement of security functions using AI.
• Strong coding skills in Python and additional languages.
• Ability to manage complex AI/security transformation programmes and engage effectively at executive levels.