AI Security Researcher: Risk & Governance for Safe AI

Belsoftsolutions; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

Belsoft is seeking a security researcher to study AI agents, assess failure modes, and translate findings into practical risk guidance for clients. You’ll collaborate with our security and AI engineers to test defenses, write risk assessments, and shape governance policies.

You will track incidents, publish research, and map deployments to OWASP, NIST, and EU AI Act frameworks, while articulating complex risk to non-technical leaders.

Qualifications

• Experience in security research, threat intelligence, or related analytical field.
• Solid understanding of how LLMs, AI agents, and tool-connected systems work.
• Familiarity with security frameworks (OWASP, NIST, ISO 27001).
• Excellent writing: translate complex risk into guidance for leaders.
• Rigorous and curious with evidence-based judgement.
• Ability to read technical material and collaborate with engineers.

Responsibilities

• Research AI security threats: prompt injection, data leakage, agent and tool abuse, supply-chain risk, and new attack techniques.
• Track incidents, disclosures, papers, and standards; highlight what matters for clients.
• Assess AI deployments at design level: access, trust boundaries, and potential failures.
• Write risk assessments, checklists, and governance policies for client teams.
• Map AI use against OWASP Top 10, NIST RMF, and EU AI Act frameworks.
• Brief engineers on tests needed and translate findings into tool requirements (DeployReady).
• Publish research: reports, articles, and talks that explain AI risk clearly.

Skills

AI security

Threat intelligence

Risk analysis

Technical writing

Analytical reasoning

Non-technical communication

Job description

Belsoft is seeking a security researcher to study AI agents, assess failure modes, and translate findings into practical risk guidance for clients. You’ll collaborate with our security and AI engineers to test defenses, write risk assessments, and shape governance policies.

You will track incidents, publish research, and map deployments to OWASP, NIST, and EU AI Act frameworks, while articulating complex risk to non-technical leaders.