AI Security Engineer

Fisher & Phillips LLP•Atlanta, GA

6d•Hybrid

About The Position

Fisher Phillips is seeking an experienced AI Security Engineer to design, implement, and continuously improve security controls for enterprise artificial intelligence, machine learning, and generative AI systems. The role secures AI platforms, models, data pipelines, integrations, agents, and supporting cloud and SaaS environments throughout the AI lifecycle. The engineer serves as a hands-on security specialist and technical partner to cybersecurity, IT, data, application, privacy, compliance, and business teams, applying security engineering, AI/ML, cloud security, secure development, threat modeling, and risk-based decision-making. The position is based in Atlanta and uses a hybrid work arrangement. Candidates must reside in the Atlanta area. Relocation costs are not covered.

Requirements

Bachelor's degree in Cybersecurity, Computer Science, Information Technology, Engineering, or a related field, or equivalent experience.
At least 5 years of security engineering, application/product security, cloud security, software security, or related cybersecurity experience.
Hands-on experience securing generative AI and LLM systems, including RAG, agents, model/API integrations, and enterprise AI platforms.
Experience with threat modeling, secure design/SDLC, application/API security testing, IAM, secrets management, and security logging/monitoring.
Proficiency in Python or another programming or scripting language.
Knowledge of APIs, automation, CI/CD, or security tooling.
Knowledge of AI threats and mitigations, including prompt injection, data leakage/poisoning, model extraction, insecure tool use, excessive agency, and AI supply-chain risk.
Familiarity with cloud AI services.
Must reside in the Atlanta area.

Nice To Haves

Experience with AI red teaming, adversarial testing, abuse-case analysis, or security evaluations of LLM, RAG, and agentic systems.
Experience with vector databases, model gateways, LangChain/LlamaIndex, MCP, and major cloud AI services such as Azure OpenAI, AWS Bedrock, or Google Vertex AI.
Familiarity with the OWASP Top 10 for LLM Applications, MITRE ATLAS, and NIST AI RMF.
Experience with IaC, policy-as-code, containers, or CI/CD security.
Relevant security, cloud, or AI certifications.
Experience in a law firm, professional services, regulated industry, or client-service environment.

Responsibilities

Design and maintain security architecture for generative AI, machine learning, AI-enabled SaaS applications, agents, APIs, and supporting cloud services.
Establish security standards, reference patterns, and guardrails for AI development, deployment, integration, and use.
Perform security reviews and threat modeling for AI solutions, including model access, data flows, retrieval-augmented generation, plugins, tools, agents, and third-party integrations.
Evaluate AI technologies, models, vendors, and services to incorporate security and privacy requirements into design and implementation.
Implement controls to protect AI platforms, model endpoints, APIs, orchestration layers, vector databases, development environments, and supporting cloud infrastructure.
Assess and reduce risks such as prompt injection, sensitive data exposure, insecure tool use, excessive agency, model supply-chain compromise, and unauthorized AI usage.
Partner with application and platform teams to embed secure development, testing, secrets management, logging, and vulnerability remediation into AI solutions.
Apply least privilege, strong authentication, service identity, privileged access, and lifecycle controls to AI platforms, models, agents, and workloads.
Protect training, grounding, prompt, output, and telemetry data using data classification, encryption, DLP, access controls, and appropriate retention practices.
Review model and agent permissions, data access, connectors, and tool integrations to identify excessive privileges, unsafe trust relationships, and unintended data exposure.
Implement and optimize controls for AI systems, including model gateways, content filtering, runtime guardrails, API security, cloud security, SIEM, DLP, and monitoring.
Develop detections and response procedures for AI-related misuse, policy violations, data leakage, anomalous model activity, and compromised integrations.
Automate security validation, testing, monitoring, and enforcement using scripting, APIs, CI/CD, infrastructure-as-code, or orchestration tools.
Partner with compliance, privacy, legal, data, and business stakeholders to align AI security controls with policies, client expectations, governance requirements, and organizational risk objectives.
Document AI security standards, architecture decisions, threat models, control designs, risk exceptions, and remediation plans.
Support AI risk assessments, incident investigations, tabletop exercises, and reviews of emerging AI threats, vulnerabilities, and attack techniques.
Communicate AI security risks and recommendations clearly to technical and non-technical stakeholders.