Position Overview

We are seeking a senior-level AI/ML Security Engineer to help establish and mature security practices across machine learning, generative AI, and agentic development environments.
This is a Hybrid onsite engagement, located in NJ 3 days a week, ideally paying on a W2 hourly basis.

This role sits at the intersection of cybersecurity, software engineering, MLOps, and emerging AI technologies. The engineer will be responsible for identifying risks within AI/ML systems and building practical security controls into the platforms, pipelines, and development processes used to create and deploy them.

A major focus will be securing the rapidly evolving use of LLMs, AI coding assistants, autonomous agents, RAG architectures, and agentic software-development workflows. The ideal candidate is highly technical and capable of both defining security strategy and implementing controls through code and automation.

Key Responsibilities

• Design and implement security controls throughout AI/ML and MLOps pipelines, including data ingestion, model development, validation, storage, deployment, and inference.

• Assess security risks associated with machine learning models, LLM applications, AI agents, datasets, prompts, and AI-enabled applications.

• Evaluate the security of AI coding assistants, coding agents, autonomous agents, and agentic workflows used within software engineering organizations.

• Establish security guardrails for AI-assisted development platforms, agent orchestration frameworks, and autonomous development pipelines.

• Help develop an enterprise framework for securing the Agentic Development Lifecycle (ADLC), incorporating threat modeling, secure development requirements, testing, deployment controls, approvals, and continuous monitoring.

• Evaluate AI Security Posture Management (AI-SPM) capabilities and establish processes for discovering, classifying, prioritizing, and remediating AI-related security risks.

• Assess emerging attacks against LLMs, frontier AI models, and autonomous agents and translate those risks into preventative and detective security controls.

• Develop and maintain automation and security capabilities using Python and CI/CD technologies.

• Integrate model and AI security scanning into development pipelines as part of a shift-left security strategy.

• Analyze model-scanning and vulnerability-assessment results and partner with engineering teams on remediation.

• Assess model inference and deployment architectures with consideration for security, performance, scalability, and resource utilization.

• Evaluate security surrounding agent sandboxes, runtime environments, tool access, permissions, agent-to-tool communication, and execution workflows.

• Establish controls governing autonomous agent behavior, including permissions, approval mechanisms, runtime restrictions, secrets management, and data-access boundaries.

• Partner closely with application security, platform engineering, software development, data science, and machine learning teams.

• Research emerging AI security threats and recommend improvements to enterprise security architecture and engineering practices.

Required Experience

• 8+ years of experience across software engineering, cybersecurity, application security, platform engineering, or related technical disciplines.

• 5+ years of hands-on software engineering or development experience.

• Strong understanding of AI/ML security, GenAI security, and agentic AI risks.

• Hands-on experience building or supporting MLOps pipelines and model deployment environments.

• Experience with platforms such as MLflow, Kubeflow, AWS SageMaker, or comparable MLOps technologies.

• Strong Python programming and automation skills.

• Strong understanding of modern CI/CD pipelines and secure software-development practices.

• Experience incorporating security testing or scanning into automated development pipelines.

• Hands-on familiarity with AI-assisted development tools such as GitHub Copilot, Claude Code, Cursor, Windsurf, Microsoft Copilot, or similar platforms.

• Experience using AI-assisted engineering techniques across one or more languages such as Python, Java, JavaScript, C#, .NET, or Go.

• Strong understanding of LLMs, AI agents, autonomous workflows, RAG architectures, tool-calling systems, and agent orchestration.

• Experience assessing the security implications of agent runtime environments, sandboxing, tool permissions, and autonomous execution.

• Ability to assess and prioritize risks involving AI models, prompts, datasets, agents, and AI-enabled applications.

• Strong understanding of AI/ML attack vectors, including:

  • Prompt injection

  • Data and model poisoning

  • Model extraction and inversion

  • Adversarial inputs and examples

  • AI/ML supply-chain vulnerabilities

  • Excessive agent permissions and unsafe tool usage

  • Sensitive-data exposure

• Familiarity with industry guidance such as the OWASP security frameworks for LLM and machine-learning applications.

• Experience with model vulnerability scanning, model security assessment, or similar AI security tooling.

• Understanding of common ML model and serialization formats such as Pickle, TensorFlow formats, and SafeTensors.

• Familiarity with both structured and unstructured data environments, including SQL databases, data warehouses, object storage, and NoSQL platforms.

• Understanding of cloud, container, microservices, and application security principles.

• Excellent analytical and problem-solving skills.

Technical Environment

Experience with several of the following technologies would be valuable:

AI/ML & Data

• MLflow

• Kubeflow

• SageMaker

• Databricks

• RAG architectures

• LLM and agent frameworks

Development

• Python

• Java

• C# / .NET

• JavaScript

• Go

• REST APIs

• Microservices

DevSecOps / CI/CD

• Git

• Bitbucket

• Jenkins

• Artifactory

• Nexus

• SonarQube

• Snyk

• Jira

• Automated security scanning

Preferred Background

• Previous experience as a software engineer or software architect before moving into security.

• Experience implementing model-scanning capabilities across an enterprise ML development environment.

• Experience defining security architecture for GenAI or agentic AI platforms.

• Familiarity with AI Security Posture Management concepts and tooling.

• Strong knowledge of secure design patterns for cloud-native and containerized applications.

• Experience working within Agile engineering organizations.

• Experience communicating emerging technical risks to engineering leadership and non-security stakeholders.

Education & Certifications

• Bachelor's degree in Computer Science, Cybersecurity, Computer Engineering, Information Systems, or a related technical discipline, or equivalent professional experience.
• Security certifications such as CISSP, CSSLP, CEH, GCIA, GPEN, or GWAPT are beneficial but not required.