Software Engineer II, Perception Red Agent Team

What you'll do

• Lead the design and architecture for secure, reliable, and safe agentic red teaming capabilities at enterprise scale.
• Produce extensible, maintainable, well-tested, secure, and performant code adhering to design specifications.
• Design, implement, test, roll out, and run software services for multi-agent systems.
• Create clear testing strategies to ensure product quality and prevent regression in existing code.
• Review product and test code to ensure it meets team standards and contains correct test coverage.
• Use debugging tools, logs, and telemetry to verify assumptions and identify issues before production.
• Convert emerging adversary tradecraft and industry frameworks into scalable, customer-ready agent capabilities.
• Ship code to thousands of machines serving millions of users globally.

What we're looking for

• Bachelor's Degree in Computer Science or related technical field and 2+ years of technical engineering experience with coding in languages like C, C++, C#, Rust, Java, JavaScript, or Python (or equivalent experience).
• Ability to meet Microsoft, customer, and/or government security screening requirements, including a Microsoft Cloud background check every two years.
• Knowledge of using advanced tooling or AI to improve work efficiency (preferred).
• Experience in software development with the ability to produce maintainable and well-tested code (preferred).
• Understanding of algorithms, data structures, and systems architecture factors affecting code quality, performance, and customer experience (preferred).
• Proficiency with debugging, testing, and telemetry tools (preferred).
• Experience building Generative AI or agentic systems, including LLM orchestration, retrieval, tool use, evaluation, and safety guardrails (preferred).
• Background in offensive security, red teaming, penetration testing, or adversary emulation, including familiarity with frameworks such as MITRE ATT&CK (preferred).