Red Team Engineer

Talent Smart Limited; Cubley; On-site; GBP 117,000 - 159,000; Full time

Job summary

Talent Smart Limited seeks an experienced Adversarial Tooling Engineer to join a major cybersecurity programme on a 14-month contract. The role focuses on designing, building and maintaining internal tooling for Red Team and adversarial security, with a significant AI component.

You will develop AI-enabled tooling and agentic workflows, implement guardrails and audit trails, and collaborate with adversarial engineers, defensive security teams and platform specialists.

Qualifications

• Strong software engineering background with production tooling experience.
• Offensive security experience in Red Team or adversary simulation.
• Strong Python development skills.
• Experience with AI-enabled tooling and AI agents or agentic workflows.
• Understanding of LLM security and autonomous AI risks.
• Knowledge of offensive security TTPs and secure tooling practices.

Responsibilities

• Design and build internal adversarial tooling, scripts, and services.
• Translate Red Team requirements into reliable technical solutions.
• Develop operator-facing tooling for real security assessments.
• Build AI-enabled tooling and agentic workflows to accelerate activities.
• Implement AI guardrails, approvals, environment controls and audit trails.
• Develop testing and regression suites for AI capabilities.
• Ensure tooling operates in constrained environments with limited permissions.
• Improve tooling via CI/CD, testing and release processes.
• Produce documentation and runbooks; support time-sensitive engagements.

Skills

Python development

Go

C#

C++

TypeScript

Linux

CI/CD

Security tooling

Red Team tooling

Tools

Git

CI/CD pipelines

AWS

Azure

Job description

Adversarial Tooling Engineer - AI & Red Team Engineering - Contract

Rate

£600 per day - Inside IR35

Contract

14 month contract with occasional travel to Sheffield

Role Overview

We are looking for an experienced Adversarial Tooling Engineer to join a major cybersecurity programme within a large, complex enterprise environment.

This is not a traditional penetration testing role. We are looking for a strong software engineer with genuine offensive security experience who can design, build and maintain the internal tooling used by Red Team and adversarial security specialists.

The role also has a significant AI component. You will develop AI-enabled adversarial tooling and agentic workflows that improve operator effectiveness while ensuring activity remains safe, controlled, auditable and within agreed rules of engagement.

What You'll Be Doing

• Design and build internal adversarial security tools, scripts, utilities and small services.
• Translate Red Team and adversarial TTP requirements into reliable technical solutions.
• Develop operator-facing tooling capable of performing effectively during real security assessments.
• Build AI-enabled tooling and agentic workflows to automate and accelerate repeatable adversarial activities.
• Implement appropriate AI guardrails including action allow-lists, human approval gates, environment controls and audit trails.
• Develop evaluation and regression testing for AI and agentic capabilities.
• Build tooling capable of operating in constrained environments with intermittent connectivity, limited permissions and challenging operational conditions.
• Apply secure-by-design principles including least privilege, secrets management, input validation and safe defaults.
• Improve existing tooling through refactoring, automated testing, CI/CD and better release processes.
• Produce high-quality documentation, runbooks and troubleshooting guidance.
• Support Red Team operators during time-sensitive engagements.
• Work closely with adversarial engineers, defensive security teams and platform specialists.
• Mentor less experienced engineers and contribute to engineering standards across the team.

What We're Looking For

• Strong software engineering background with experience building maintainable production-quality tools or services.
• Practical offensive security experience across Red Teaming, penetration testing or adversary simulation.
• Strong Python development skills.
• Experience with additional languages such as Go, C#, C++ or TypeScript.
• Experience building AI-enabled tooling, AI agents or agentic workflows.
• Understanding of LLM security and the risks associated with autonomous or semi-autonomous AI systems.
• Strong understanding of offensive security TTPs and operational security.
• Experience developing secure tooling with appropriate credential handling, logging and traceability.
• Strong Linux experience.
• CI/CD, automated testing and modern software engineering practices.
• Ability to operate effectively within defined security governance and rules of engagement.

Desirable Experience

• Development of Red Team/operator-facing CLI or GUI tooling.
• Development of APIs, automation frameworks or agent/implant-side components.
• AWS and/or Azure.
• Defensive security, detection engineering or Purple Team experience.
• AI security including prompt-injection defence, agent permissions, tool/action controls and human-in-the-loop workflows.

The Ideal Candidate

You are likely to come from a strong software engineering background before moving deeper into offensive security, or you may already be an experienced Red Team engineer who has built substantial tooling rather than simply used commercial or open-source security products.

Crucially, you should be someone who enjoys building things. We are looking for an engineer capable of taking an operator requirement and turning it into robust, reusable and maintainable adversarial capability.

Experience combining offensive security engineering with AI or agentic technology would be particularly attractive.

Why Consider This Contract?

This is an opportunity to work at the intersection of software engineering, offensive cybersecurity and AI. Rather than simply operating existing security tools, you will help build the capabilities that specialist adversarial teams use, including the next generation of AI-assisted security tooling.

14 month contract with occasional travel to Sheffield