DESE's Cyber Works and Digital Engineering teams design, build, and integrate emerging AI/ML technologies to harden and secure the systems that defend the nation, across ground, missile defense, space, and installation infrastructure S&T programs. We're expanding our Secure AI practice to build and assure trusted, robust AI/ML solutions that interpret complex datasets, predict outcomes, and automate decision-making in support of critical military platforms.
This is a consolidated announcement covering multiple tracks; your assignment may emphasize one or a blend of: Classic ML & Predictive Modeling, LLM/GenAI Applications, AI Assurance & Responsible AI, Edge AI/ML Deployment, and hardening AI/ML systems against adversarial threats. All tracks require independent research, cross-functional collaboration, and the ability to clearly communicate complex technical work to stakeholders.
Core Responsibilities:- Design, develop, and maintain machine learning models, from classical algorithms (regression, tree ensembles, clustering) to modern deep learning architectures.
- Build LLM-enabled applications and retrieval-augmented generation (RAG) pipelines, including vector embedding generation, vector database integration, and prompt/context engineering.
- Develop AI assurance and evaluation tooling: robustness testing, bias/fairness analysis, model traceability, red-team/adversarial testing, and audit artifact generation.
- Optimize and deploy models for production and edge environments (quantization, compression, containerized inference, ONNX/TensorRT).
- Implement secure model and data pipelines, defend against adversarial ML threats, and ensure supply-chain integrity (SBOM) for AI components.
- Conduct data processing/analysis to improve model accuracy; document and present development processes and assurance evidence to stakeholders.
- Contribute to Agile, team-based planning and estimating in a fast-paced, collaborative environment.
Minimum Requirements: - Bachelor's degree or equivalent experience in CS/CPE/EE/Data Science (or related field).
- Proven experience in one or more: ML/LLM development, assurance/evaluation tooling, or deploying edge computing solutions.
- Hands-on experience with ML frameworks such as TensorFlow, PyTorch, or Scikit-learn.
- Proficiency in Python and at least one additional language (Java, C++).
- Strong understanding of data structures, data modeling, and software architecture.
Highlighted Skills & Experience:- Modern AI: LLM frameworks and application development; vector embeddings and vector databases; RAG architectures; prompt/context engineering; LLM fine-tuning; model evaluation and benchmarking.
- Classic ML: Feature engineering, model selection/tuning, statistical analysis, and predictive modeling across structured and unstructured data.
- AI Assurance: Responsible AI practices (robustness, bias/fairness, traceability); adversarial ML defenses; red-team testing; auditability and compliance documentation.
- Edge & Deployment: ONNX/TensorRT, quantization/compression, ARM/NVIDIA Jetson/DSP targets, containerized inference, MLOps in controlled/classified environments.
- Platform & DevSecOps: REST APIs, CI/CD for software and ML systems, SAST/DAST, Infrastructure-as-Code, cloud platforms (AWS, Azure, GCP).
- Domain: Familiarity with computer networking, secure system integration for mission platforms, and test/V&V support (Python/MATLAB analysis).
- Contributions to open-source AI/ML projects; experience deploying AI models in production, classified, or embedded environments.
- Understanding of computer security principles and secure software development lifecycle (SSDLC) practices.
Why This Role Matters:You'll join a high-impact team solving some of the DoD's hardest problems in AI-enabled system security, building the next generation of trusted, auditable, and mission-ready AI for national defense.
