Enterprise Cybersecurity AI Infrastructure Engineer

Booz Allen Hamilton•McLean, VA

1d•$62,000 - $141,000

About The Position

The Enterprise Cybersecurity AI Infrastructure Engineer will operate as a hands-on technical contributor responsible for building, scaling, and operationalizing Booz Allen's Cyber Operations AI infrastructure platform. This role supports the training, evaluation, hosting, and serving of AI capabilities. The engineer will lead key components of the platform, including Databricks for data and model delivery and GitHub with AWS for the deployment pathway of internal AI products into production. Responsibilities include streamlining collaboration between development and operations, ensuring efficient systems and workflows, and working with AI/ML engineers, data scientists, and security operators to transform security telemetry, threat intelligence, and analyst workflows into reliable, scalable pipelines and services. The role also involves building and operating CI/CD and DevSecOps for these products, including GitHub Actions, infrastructure as code, identity, secrets, and least-privilege access in Unity Catalog. The engineer will perform design and code reviews, set technical direction for platform work, and translate researcher and analyst needs into reusable infrastructure, leveraging strong software engineering and communication skills to assess platform problems, align stakeholders, and drive decisions to closure.

Requirements

2+ years of experience with software engineering including building and operating production-scale machine learning infrastructure, data platforms, or distributed systems
2+ years of experience creating and training agents and models for agentic AI, including tool calling, evaluation, and deployment into production workflows
Experience with cloud platforms such as AWS and Databricks and building CI/CD in GitHub Actions with AWS integration, including IAM, OIDC federation, and Secrets Manager
Experience in Python and SQL, and working with infrastructure as code including Terraform or Terragrunt while also utilizing coding agents
Experience building and maintaining data pipelines over large or sensitive datasets, including schema management and reproducible deployments
Experience owning platform work end to end, from technical design through production deployment and operational monitoring
Experience applying DevSecOps practices to delivery pipelines, including least-privilege access, secret handling, and automated security checks
Knowledge of cybersecurity operations or security telemetry
Ability to obtain a Secret clearance
HS diploma or GED

Nice To Haves

Experience with containerization such as Docker and container orchestration such as Kubernetes
Experience with Spark, distributed training, or orchestration frameworks such as Ray or PyTorch Distributed
Experience with MLflow or an equivalent stack for experiment tracking, evaluation harnesses, and a model registry on Databricks
Experience implementing retrieval and agent infrastructure, including vector indexes, model endpoints, tool gateways, and evaluation loops
Experience with AWS beyond identity and secrets across storage, networking, or private connectivity into Databricks
Experience operating platforms in sensitive or regulated environments including FedRAMP, IL4, or IL5
Knowledge of secure-AI practices aligned to NIST AI RMF, OWASP LLM Top 10, and MITRE ATLAS
Knowledge of SIEM, SOAR, or threat-intelligence platforms and how models incorporate into analyst workflows
IT, cloud, or security Certification such as AWS, Databricks, CKA, CKAD, CISSP, or an AI security Certification

Responsibilities

Partner with AI and ML engineers, data scientists, and SOC operators to turn modeling and analyst requirements into shared platform capabilities, including retrieval, scoring, and agent workflow design.
Build, operate, and optimize the platform behind internal AI products, including platforms based on Databricks and AWS.
Implement and scale data pipelines to land security telemetry and threat intelligence including logs, endpoint, identity, cloud, and threat-intel sources into governed schemas analysts and models can use.
Build model evaluation, validation, and serving paths, including experiment tracking, regression checks, and safe rollout and rollback, so models reach operations with a measured quality bar.
Build and maintain GitHub Actions CI/CD for AI products including self-hosted runners, AWS OIDC into IAM, Secrets Manager, pull-request environments, and production promotion.
Implement DevSecOps controls, including dependency and secret scanning, infrastructure policy, service-principal isolation, and Unity Catalog grants to keep non-production identities off production data.
Own infrastructure as code for catalogs, permissions, and deployment such as Terraform or Terragrunt and Databricks Asset Bundles so every environment is repeatable.
Use coding assistants to accelerate delivery, and review generated infrastructure, pipeline, and application changes before they merge.