Software Engineer II, Attack Detection

Abnormal

17h•$149,200 - $214,500

About The Position

At Abnormal AI, the mission is to safeguard large enterprises against evolving email attacks using a behavioral-based approach. The company’s system protects more than 15% of the Fortune 500. The Software Engineer II builds customer-facing security features and supporting software, including detection services, AI agents for investigating suspicious activity, and systems that support model development and evaluation through automated agentic loops. The engineer works closely with ML Engineers, Data Scientists, and Software Engineers to turn new approaches into reliable production systems. The work aims to improve attack detection and security investigations while limiting false positives and cost, and to maintain a world-class inbound email security product. Strong software engineering skills are essential; practical machine learning experience is preferred.

Requirements

Bachelor's degree in Computer Science, Software Engineering, Information Systems, or a related engineering field.
At least 2 years of experience in software design and development.
Experience with backend services or data-intensive applications where reliability and performance matter.
Strong programming skills in a general-purpose language and knowledge of data structures and software design.
Experience with automated tests and diagnosis of production issues through logs and metrics.
Ability to turn a product requirement into a practical design and deliver it with guidance on larger technical decisions.
Clear technical communication and effective collaboration with other engineers.

Nice To Haves

Hands-on machine learning experience from industry, research, or substantial projects.
Experience with large language models or AI agents, including evaluation of accuracy and reliability.
Experience with statistical evaluation of classifiers, including precision and recall.
Experience with Python or Go.
Experience with data pipelines or tools that make model experiments reproducible.
Experience with cybersecurity or fraud detection.

Responsibilities

Build backend services and data pipelines for customer-facing security features.
Integrate machine learning models and large language models into detection and investigation workflows.
Develop AI agents that analyze suspicious activity and support security investigations.
Build tools that make model training repeatable and agent-driven, and help engineers evaluate AI systems.
Evaluate changes against real attack patterns and investigate missed attacks and false positives.
Improve system reliability and performance within latency and cost limits.
Write tested, readable code and contribute to technical designs and code reviews.
Own well-defined features from design through release and monitor their behavior in production.