Gen AI Security Architect

Envision Technology Solutions; United States; On-site; USD 120,000 - 170,000; Full time

Job summary

Envision Technology Solutions seeks a seasoned AI Security Engineer to design and implement comprehensive AI security controls across AI/ML lifecycles. You will develop prompt security strategies, enforce data protection, and lead threat modeling for LLM applications.

You will collaborate with engineering, cybersecurity, data, and compliance teams to ensure guardrails, policy compliance, and incident response readiness in enterprise environments.

Qualifications

• Strong understanding of Generative AI, LLMs, AI agents, and prompt engineering.
• Experience with AI threat modeling and AI security controls.
• Strong knowledge of prompt injection, jailbreaks, adversarial prompts, and LLM threats.
• Experience designing and implementing AI guardrails and content-safety controls.
• Knowledge of data security, DLP, encryption, access control, and data classification.
• Understanding IAM and secure API design.
• Experience with LLM input/output filtering, validation, monitoring, and security testing.
• Familiarity with OWASP Top 10 for LLM Applications and MITRE ATLAS.
• Knowledge of secure software development and cloud security principles.
• Experience with security logging, monitoring, vulnerability assessment, and incident response.
• Strong understanding of privacy, compliance, and responsible AI principles.

Responsibilities

• Design and implement AI security controls and guardrails for AI apps.
• Develop and maintain prompt security strategies including prompt injection and adversarial prompts.
• Implement controls for sensitive data protection (PII, confidential data).
• Design security mechanisms for LLM inputs, outputs, context, and data flows.
• Develop and enforce AI guardrails for content safety and access control.
• Conduct AI threat modeling and risk assessments for LLM apps.
• Establish controls for prompt and response validation, filtering, monitoring, and logging.
• Identify and mitigate risks such as data exfiltration and model abuse.
• Collaborate with engineering, cybersecurity, data, and compliance teams across the AI lifecycle.
• Define security policies for responsible use and deployment of generative AI.
• Monitor AI systems for incidents and emerging attack patterns; support incident response.
• Perform security testing and red-team assessments of AI applications and guardrails.
• Evaluate AI security tools and frameworks; recommend improvements to security architecture.
• Maintain documentation covering AI security policies, threat models, controls, and risk assessments.

Skills

Generative AI

LLMs

AI agents

prompt engineering

AI threat modeling

AI security controls

Prompt injection

Jailbreaks

Adversarial prompts

LLM threats

AI guardrails

Content-safety controls

Data security

DLP

Encryption

Access control

Data classification

IAM

Secure API design

Input/output filtering

Validation

Monitoring

Security testing

OWASP Top 10 for LLM

MITRE ATLAS

Secure SDLC

Cloud security

Security logging

Vulnerability assessment

Incident response

Privacy

Compliance

Responsible AI

OpenAI platforms

Azure OpenAI

AWS

Google Cloud

Guardrail platforms

LLM security testing tools

APIs

CI/CD

DevSecOps

RAG architectures

Vector databases

Tool security

Enterprise AI governance

Tools

OpenAI platforms

Azure OpenAI

AWS

Google Cloud

Guardrail platforms

LLM security testing tools

APIs

CI/CD

DevSecOps

RAG architectures

Vector databases

Tool/function calling security

AI governance standards

Job description

• Design and implement AI security controls and guardrails to protect AI applications from misuse, abuse, and emerging threats.
• Develop and maintain prompt security strategies, including prompt injection, jailbreak, instruction manipulation, and adversarial prompt detection.
• Implement controls for sensitive data protection, including PII, confidential information, and proprietary business data.
• Design security mechanisms for LLM inputs, outputs, context, and data flows.
• Develop and enforce AI guardrails for content safety, data leakage prevention, access control, and policy compliance.
• Conduct AI threat modeling and risk assessments for LLM applications and AI agents.
• Establish controls for prompt and response validation, filtering, monitoring, and logging.
• Identify and mitigate risks such as data exfiltration, prompt injection, model abuse, unauthorized access, and unsafe model behavior.
• Work with engineering, cybersecurity, data, and compliance teams to integrate AI security controls into the AI/ML development lifecycle.
• Define security policies and standards for the responsible use and deployment of generative AI.
• Monitor AI systems for security incidents and emerging attack patterns and support incident response when required.
• Perform security testing and red-team assessments of AI applications and guardrail implementations.
• Evaluate AI security tools, frameworks, and technologies and recommend improvements to the organization's AI security architecture.
• Maintain documentation covering AI security policies, threat models, controls, guardrails, and risk assessments.

Required Skills

• Strong understanding of Generative AI, LLMs, AI agents, and prompt engineering.
• Experience with AI threat modeling and AI security controls.
• Strong knowledge of prompt injection, jailbreaks, adversarial prompts, and LLM-specific threats.
• Experience designing and implementing AI guardrails and content-safety controls.
• Knowledge of data security, data loss prevention (DLP), encryption, access control, and data classification.
• Understanding of identity and access management (IAM) and secure API design.
• Experience with LLM input/output filtering, validation, monitoring, and security testing.
• Familiarity with AI security frameworks such as the OWASP Top 10 for LLM Applications and MITRE ATLAS.
• Knowledge of secure software development and cloud security principles.
• Experience with security logging, monitoring, vulnerability assessment, and incident response.
• Strong understanding of privacy, compliance, and responsible AI principles.

Preferred Experience

• Experience securing OpenAI, Azure OpenAI, AWS, Google Cloud, or other enterprise AI platforms.
• Experience with AI security/guardrail platforms and LLM security testing tools.
• Experience working with APIs, cloud platforms, CI/CD pipelines, and DevSecOps.
• Familiarity with RAG architectures, vector databases, AI agents, and tool/function calling security.
• Experience developing AI security policies and enterprise AI governance standards.

AI Security | Prompt Security | Threat Modeling | LLM Security | AI Guardrails | Data Security | Data Privacy | DLP | Prompt Injection Defense | AI Risk Management | Security Testing | Responsible AI | Cloud Security | DevSecOps