Research Staff, Adversarial Intelligence

Horizon3

5d•$145,000 - $170,000•Remote

About The Position

Horizon3.ai is seeking a Research Staff Member to join its AI Research team and investigate research problems at the intersection of super intelligence, systems security, and agentic systems for security. The role focuses on technical problems arising from real security environments and turning research ideas into methods, prototypes, experiments, and systems that can influence Horizon3's technology. Projects may involve AI agents for security, autonomous penetration testing, attack-path reasoning, cyber deception, AI safety and security, evaluation of AI systems, or methods for reasoning over security data and environments. This is a research-and-engineering role rather than a purely academic or conventional software engineering position. The successful candidate will have strong computer science foundations and meaningful depth in at least one of AI or systems security, and be motivated to develop expertise in the other. The position may be hired at P1 (Entry) or P2 (Developing), depending on experience and demonstrated technical maturity. Junior researchers will receive mentorship and technical guidance as they develop independence. Horizon3.ai is a remote cybersecurity company whose NodeZero platform provides production-safe autonomous penetration testing and other assessment operations across internal, external, cloud, and hybrid cloud environments. The company describes its culture as collaborative and grounded in respect, ownership, and results. This job is fully remote and may require up to 5% travel. The stated annual base salary range is $145,000–$170,000, with the exact amount determined by location, qualifications, experience, and relevant skills. Full-time roles are eligible for stock options. Listed benefits include health, vision, and dental insurance for employees and their families, a flexible vacation policy, and generous parental leave.

Requirements

Strong foundations in computer science, including the ability to reason about algorithms, systems, software, and data.
Technical depth in systems security, super intelligence or machine learning, or a closely related area, and strong interest in developing expertise across AI and security.
Strong programming skills and ability to translate technical ideas into working implementations; Python is particularly useful, but no specific language is required.
Ability to design careful experiments, analyze results, and distinguish evidence from intuition.
Curiosity and willingness to investigate unfamiliar technical areas in depth.
Interest in both research and engineering, with comfort moving between technical ideas, experiments, prototypes, and implementation.
Ability to receive technical feedback, revise approaches, and learn from researchers and engineers.
Good technical judgment, including careful reasoning about assumptions, limitations, risks, and unintended consequences.
Strong written and verbal communication skills.
A degree in Computer Science or a related field, or equivalent technical experience.

Nice To Haves

Experience in systems security, offensive security, defensive security, penetration testing, vulnerability research, operating systems, networks, distributed systems, cloud security, or related areas.
Experience with machine learning, deep learning, large language models, AI agents, reinforcement learning, planning, search, graph-based reasoning, or related AI methods.
Experience conducting research through an academic research group, thesis, independent research project, internship, industrial research project, or comparable technical work.
Experience designing experimental infrastructure, benchmarks, test environments, or evaluation frameworks.
Experience building systems that operate on real-world data, environments, or security infrastructure rather than only clean benchmark datasets.
Familiarity with security concepts such as attack paths, Active Directory, cloud environments, identity systems, network security, security controls, or adversarial techniques.
Strong research publications, conference presentations, open-source contributions, or other evidence of high-quality technical work; publications are valuable but not required.
Experience working across research and engineering teams or contributing research ideas to practical systems.

Responsibilities

Conduct research on problems at the intersection of AI, cybersecurity, and computer systems.
Formulate research questions based on technical challenges faced by Horizon3 and its customers.
Design and run experiments to test research hypotheses in controlled and representative environments.
Build research prototypes and implement methods at a quality level that supports meaningful evaluation.
Study and evaluate modern AI techniques, including large language models, agentic systems, reasoning methods, and learning algorithms, as relevant to research problems.
Analyze security systems, attack paths, telemetry, graphs, configurations, and other real-world security data.
Read and critically evaluate academic and technical literature, identify relevant ideas, and assess their application to Horizon3's research problems.
Collaborate with researchers, attack engineers, security engineers, product teams, and other technical groups to understand problems and evaluate potential solutions.
Document research methods, experiments, results, and limitations clearly and rigorously.
Communicate findings through internal presentations, technical reports, prototypes, and, when appropriate, academic publications or external research outputs.
Develop greater research independence, including identifying promising questions, proposing technical approaches, and driving portions of research projects.