Security Engineer (Threat Response), Sophos Security Team (IDR)

About Us
Sophos is a cybersecurity leader defending 600,000 organizations globally with an AI-driven platform and expert-led services. Sophos meets organizations wherever they are in their security maturity and grows with them to defeat cyberattacks. Its solutions combine machine learning, automation, and real-time threat intelligence with frontline human expertise from Sophos X-Ops to deliver advanced, 24/7 threat monitoring, detection, and response.

Sophos offers industry-leading managed detection and response (MDR) alongside a comprehensive portfolio of cybersecurity technologies — including endpoint, network, email, and cloud security, extended detection and response (XDR), identity threat detection and response (ITDR), and next-gen SIEM. Together with expert advisory services, these capabilities help organizations proactively reduce risk and respond faster, with the visibility and scalability needed to stay ahead of evolving threats.

Sophos goes to market with a global partner ecosystem, including Managed Service Providers (MSPs), Managed Security Service Providers (MSSPs), resellers and distributors, marketplace integrations, and cyber risk partners, giving organizations the flexibility to choose trusted relationships when securing their business. Sophos is headquartered in Oxford, U.K. More information is available at www.sophos.com.

Role Summary
Independently detect, investigate and respond to cyber threats across Sophos and its products, while engineering the detections, automations, reusable skills and agentic workflows that scale Internal Detection and Response. The role combines incident response judgement with security engineering and the safe, measurable and observable use of AI.

What You Will Do
- Own complex investigations across the incident lifecycle, from triage and evidence collection through containment, recovery and lessons learned.

- Act as a trusted front door for security issues, coordinating response across IDR, engineering, product and business teams.

- Perform DFIR and endpoint forensics, including log, network and packet analysis, malware analysis and firewall investigation where applicable.

- Write and improve detections, playbooks, orchestrations and preventions using evidence from incidents and threat hunting.

- Build production-quality automation that removes repetitive work, improves consistency and enables analysts to operate at a higher layer.

- Design model-agnostic, reusable skills and workflows that run across approved AI and agent platforms rather than depending on one model.

- Use Claude, Codex, Copilot and future approved tooling to accelerate investigations, coding, testing, documentation and detection engineering.

- Contribute through GitHub: create branches, raise and review pull requests, write tests, respond to feedback and support controlled deployment.

- Instrument agentic workflows with logs, traces, metrics, evaluation results and failure signals so behaviour and outcomes are observable.

- Apply safe harness design, least privilege, scoped tool access, approval gates, evidence validation, rollback and human review for consequential actions.

- Communicate incidents, risks, decisions and technical outcomes clearly to engineering and leadership audiences.

What You Will Bring
- Strong enterprise incident response experience across endpoint, identity, cloud, network and product-focused investigations.

- Practical capability in DFIR, endpoint and firewall forensics, threat hunting, detection engineering and analysis of unstructured telemetry.

- Ability to write reliable automation in Python or a comparable language, work with APIs and query data using SQL.

- Working knowledge of Git and GitHub engineering practices, including pull requests, code review, testing and CI/CD concepts.

- Hands-on familiarity with Claude, Codex or comparable coding agents, including context design, task decomposition and validation of generated output.

- Ability to design and write evals for agentic workflows and skills, covering task success, output quality, safety, regressions and common failure modes.

- Understanding of agent architecture, tool use, skill-based design, model portability and AI or automation safety controls.

- Ability to design observability for automated work and use telemetry to troubleshoot, measure quality and improve reliability.

- Clear written and verbal communication, sound opsec judgement and the ability to work effectively in a globally coordinated environment.