Note: The job is a remote job and is open to candidates in USA. Dice is hiring an AI Safety/Security Engineer for a contract W2 role. The engineer will design and operate security guardrails for LLM-based and agentic AI systems, conduct threat modeling and adversarial testing, and implement runtime monitoring, safety controls, and automated security gates for production AI applications.
Responsibilities
- Design and operate technical guardrails for LLM and agentic AI applications
- Build adversarial evaluation and **AI red-team testing harnesses**
- Develop scenario-based test suites for prompt injection, jailbreaks, RAG attacks, tool misuse, and data exfiltration
- Define AI threat models and measurable safety acceptance criteria
- Implement runtime monitoring for anomalous agent behavior
- Conduct full-scope AI security assessments and document attack paths and vulnerabilities
- Maintain regression suites for known AI attack patterns
- Integrate safety controls and automated security gates into CI/CD pipelines
- Work with Blue Teams/SOC to improve AI threat detection and incident response
- Partner with Legal, Compliance, and Responsible AI teams on regulatory requirements
- Develop audit logging and explainability mechanisms for AI safety decisions
- Present security findings and mitigation recommendations to technical and executive stakeholders
Skills
- 8+ years of experience in **Cybersecurity / AI Security / AI Safety**
- Strong experience securing **LLM-based and AI Agent systems**
- **Prompt Injection Detection, Jailbreak Resistance, PII Redaction**
- LLM **Input/Output Filtering and Safety Guardrails**
- **Tool-use constraints** and API/function access controls
- Secret/credential protection and **data exfiltration prevention**
- **AI Red Teaming / Adversarial Testing**
- Multi-turn jailbreak and indirect prompt injection testing
- Security testing of **RAG and tool-calling agents**
- AI threat modeling and risk assessment
- Runtime AI safety monitoring and automated incident response
- Logging, alerting, blocking, throttling, and forensic analysis
- CI/CD **automated safety gates and regression testing**
- Strong understanding of **Responsible AI, AI Governance, and AI Risk**
- Financial services regulatory/compliance experience
- Financial services regulatory/compliance experience preferred
Company Overview
Company H1B Sponsorship