Senior Security Engineer, Agentic AI — AI Security & Application Security

Trail of Bits is a cybersecurity research and engineering company focused on securing emerging technologies, critical infrastructure, blockchain systems, and advanced software environments. The company combines security research, custom engineering, threat modeling, and application security expertise to help organizations defend against evolving attack surfaces across modern computing stacks.

This remote full-time Senior Security Engineer, Agentic AI role sits within the Software Assurance team and focuses on application security assessments for agentic AI systems, LLM applications, inference pipelines, RAG systems, prompt injection attack surfaces, and AI deployment infrastructure. 

The position combines adversarial security research with hands-on assessments involving PyTorch, LangChain, Jax, Python, Rust, Golang, and modern AI security tooling. Candidates interested in web3 security roles, AI infrastructure security, and blockchain infrastructure roles will work across threat modeling, AI policy frameworks, prompt injection research, and security validation for production-grade agentic systems.

Operational AI security expertise is increasingly critical for organizations scaling a web3 team and deploying agentic infrastructure securely.

🔹 Responsibilities

• Conduct application security assessments across agentic AI systems, LLM pipelines, deployment infrastructure, and AI tooling ecosystems
• Analyze vulnerabilities involving model architectures, guardrails, inference systems, and deployment environments while developing mitigation strategies
• Research and operationalize prompt injection techniques targeting agentic workflows, multi-turn interactions, tool-use systems, and cross-agent attack paths
• Perform static analysis, dynamic testing, and manual code review focused on vulnerabilities intersecting application security and AI systems
• Conduct threat modeling and risk assessments involving prompt injection surfaces and orchestration-layer attack vectors
• Collaborate with client engineering teams to review architectures, codebases, and operational security controls
• Contribute to AI governance initiatives, assurance frameworks, auditing methodologies, and AI compliance processes for mission-critical deployments

Advanced AI threat modeling capabilities are increasingly important for organizations looking to build a web3 engineering team around secure AI deployment practices.

🔹 Requirements

• Demonstrated experience and interest in agentic AI security including identifying and mitigating AI-specific vulnerabilities
• Hands-on experience with prompt injection attacks and defensive mitigation techniques
• Deep understanding of AI/ML architectures, MLOps workflows, and frameworks including PyTorch, Jax, LangChain, and RAG systems
• Proven experience conducting technical application security assessments and implementing security controls
• Practical experience designing prompt injection workflows against production LLM systems and agentic environments
• Strong programming experience across languages including Rust, Golang, Kotlin, Swift, Objective-C, JavaScript/TypeScript, Python, Ruby, C, and/or C++
• Strong adversarial thinking and ability to identify novel attack vectors across multiple abstraction layers
• Strong communication skills with the ability to explain technical security findings and mitigation strategies clearly to stakeholders

🔹 Compensation & Benefits

• Base salary range: USD $100,000–$200,000 excluding bonuses and benefits
• Performance-based bonus opportunities
• Fully company-paid medical, dental, vision, disability, and life insurance coverage
• 401(k) plan with 5% salary match
• 20 days of paid vacation with additional flexibility depending on jurisdiction
• Four months of parental leave
• $10,000 relocation assistance for employees relocating to New York City
• $1,000 work-from-home stipend
• $750 annual learning and development stipend
• Company-sponsored team events including travel and accommodation
• Philanthropic donation matching up to $2,000 annually
• Remote-first work culture operating across global time zones

—
Explore more web3 security roles, blockchain infrastructure roles, and AI infrastructure security jobs on Home | ArtofBlockchain