AI Solutions Security Engineer

Company Summary

Position Responsibilities

About the Role 

We are seeking a highly motivated AI Solutions Security Engineer to join Deltek’s growing AI Center of Excellence team to design, build, and operate the security, compliance, guardrail, and policy controls that keep our internal Artificial Intelligence and Machine Learning solutions safe, trustworthy, and production-ready. The ideal candidate combines deep expertise in AI security, application and cloud security, Generative AI, Large Language Models (LLMs), SLMs, and software engineering with a practical understanding of MLOps/LLMOps to secure AI applications without slowing the teams that build them. This role reports to the Senior AI Solutions Architect. 

Key Responsibilities 

AI Security & Threat Defense 

• Design and implement security architecture and controls for LLM, Generative AI, RAG, and agentic AI applications. 

• Perform threat modeling and AI-specific risk assessments aligned to frameworks such as OWASP Top 10 for LLM Applications, MITRE ATLAS, and NIST AI RMF. 

• Defend against prompt injection, jailbreaks, data exfiltration, model and data poisoning, insecure output handling, and excessive agent autonomy. 

• Conduct AI red teaming, adversarial testing, and penetration testing of models, prompts, agents, and integrations; track findings through remediation. 

• Secure MCP servers, agent-to-agent (A2A) communication, tool calling, and plugin integrations with least-privilege access, authentication, and authorization. 

• Protect models, embeddings, vector stores, prompts, and training data with encryption, access controls, secrets management, and network isolation. 

AI Guardrails & Runtime Protection 

• Design, implement, and tune guardrails for input and output filtering, content safety, topic restrictions, grounding checks, and hallucination mitigation. 

• Build and operate PII, sensitive data, and secrets detection and redaction across prompts, retrieved context, and model responses. 

• Evaluate and integrate guardrail tooling such as AWS Bedrock Guardrails, Azure AI Content Safety, NeMo Guardrails, Guardrails AI, and LLM Guard. 

• Build automated safety, security, and bias evaluation suites and run them as quality gates in CI/CD pipelines. 

AI Compliance, Policy & Governance 

• Translate AI policies, Responsible AI principles, and regulatory requirements into enforceable technical controls and policy-as-code. 

• Support compliance with frameworks and regulations such as the EU AI Act, NIST AI RMF, ISO/IEC 42001, SOC 2, ISO 27001, GDPR, and CCPA. 

• Establish and run AI risk review, and exception processes with security, privacy, legal, and business stakeholders. 

Monitoring, Detection & Incident Response 

• Implement logging, monitoring, and alerting for AI-specific security events such as policy violations, anomalous usage, data leakage, and model abuse. 

• Integrate AI telemetry with SIEM, SOAR, and observability platforms to support detection and response. 

• Develop AI incident response playbooks and participate in triage, investigation, and post-incident reviews. 

AI Solutions Engineering 

• Collaborate with AI data scientists and solutions teams to design, build, and deploy secure Generative AI, RAG, and agentic solutions using LLMs such as GPT, Claude, Gemini, Llama, and Mistral. 

• Develop production-grade Python services, APIs, and reusable secure components, such as guardrail libraries, SDKs, and templates. 

• Integrate AI services with enterprise systems, identity providers, and data platforms using secure design patterns. 

Cloud & DevSecOps 

• Secure AI workloads on AWS (preferred) and Azure, including IAM, VPC and private networking, KMS, and secrets management. 

• Embed security scanning in CI/CD pipelines, including SAST, DAST, dependency and SBOM checks, container and image scanning, and infrastructure-as-code scanning. 

Qualifications

Required Qualifications 

Education 

• Bachelor’s or Master’s degree in Computer Science, Cybersecurity, Artificial Intelligence, Data Science, Engineering, or a related technical field. 

Experience 

• 5+ years of software engineering, application security, cloud security, or machine learning engineering experience. 

• 2+ years of hands-on experience securing, developing, or deploying Agentic AI, Generative AI, or AI/ML solutions in production environments. 

• Demonstrated experience implementing security, privacy, or compliance controls in a regulated or enterprise environment. 

Technical Skills 

AI Security, Guardrails & Governance 

• OWASP Top 10 for LLM Applications, MITRE ATLAS, and NIST AI RMF. 

• Prompt injection defense, jailbreak testing, AI red teaming, and adversarial testing techniques. 

• Guardrail and content safety tools such as AWS Bedrock Guardrails, Azure AI Content Safety, NeMo Guardrails, Guardrails AI, or LLM Guard. 

• PII and sensitive data detection and redaction, such as Microsoft Presidio, Amazon Macie, or similar tools. 

• Policy-as-code and access control tooling such as Open Policy Agent (OPA), Kyverno, or Cedar. 

• Responsible AI practices: explainability, bias mitigation, model cards, and risk management. 

Security Engineering & DevSecOps 

• Identity and access management, OAuth/OIDC, SSO, secrets management, and encryption. 

• Threat modeling, secure design review, and vulnerability management. 

• SAST, DAST, SCA, SBOM, and container and IaC scanning tools. 

• SIEM, logging, and monitoring integration for security detection and response. 

• Docker and Kubernetes; Git, GitHub, Azure DevOps, or Jenkins; CI/CD automation. 

Programming & Engineering 

• Strong expertise in Python. 

• Experience with Java, ReactJS, JavaScript, or similar programming languages. 

• Solid understanding of algorithms, data structures, APIs, and software design principles. 

Artificial Intelligence & Generative AI 

• Machine Learning and Deep Learning concepts and frameworks, including PyTorch, TensorFlow, and Scikit-learn. 

• Large Language Models (LLMs) and SLMs, prompt engineering, and Retrieval-Augmented Generation (RAG). 

• AI agents and agentic workflows using LangChain, LlamaIndex, Semantic Kernel, MCP, and A2A. 

• Fine-tuning and model customization; vector embeddings and semantic search. 

• FastAPI or Flask for building AI services. 

MLOps & Data 

• MLflow, Kubeflow, Azure ML, or SageMaker; model monitoring and lifecycle management. 

• SQL and NoSQL databases; data pipelines, ETL, and data modeling. 

Cloud Platforms 

• AWS (preferred), Azure, and Google Cloud. 

• AWS Bedrock or Azure OpenAI Service. 

• AWS SageMaker or Google Vertex AI. 

Preferred Qualifications 

• Experience designing enterprise-scale AI platforms and shared AI security services. 

• Knowledge of multi-agent architectures and autonomous AI systems, including agent identity and permission models. 

• Experience with vector databases such as Pinecone, Snowflake Cortex, Pgvector, Weaviate, Chroma, or Azure AI Search, including access control and data isolation. 

• Hands-on experience with AI governance, compliance, and Responsible AI frameworks, including ISO/IEC 42001 or EU AI Act readiness. 

Applicant Privacy Notice

• Career Interests: Engineering
• Position Type: FT
• Travel Requirements: 20%