Job description

About Agile Defense At Agile Defense we know that action defines the outcome and new challenges require new solutions. That’s why we always look to the future and embrace change with an unmovable spirit and the courage to build for what comes next. Our vision is to bring adaptive innovation to support our nation's most important missions through the seamless integration of advanced technologies, elite minds, and unparalleled agility—leveraging a foundation of speed, flexibility, and ingenuity to strengthen and protect our nation’s vital interests. Requisition #: 1921 Job Title: SME AI/ML Engineer Location: Fort Meade, MD | Onsite Clearance Level: TS/SCI w/ Polygraph, Must Have Clearance to Start Job Description Ready for a challenge that puts you at the center of AI innovation for cyber operations? Are you an experienced AI/ML engineer who thrives in a fast-paced environment, enjoys turning research and prototypes into real mission capability, and wants to help shape how AI is built and delivered across a national mission? Our team is standing up an AI Enterprise capability — the central hub that mission teams turn to when they want to move an AI solution from prototype to a secure, governed, enterprise-scale service. As a Senior AI/ML Engineer on this team, you will design and integrate the models, agents, and AI workflows that run on that platform, with significant, visible impact on the mission. What You'll Be Doing: Provide embedded operations support, capability integration, and rapid prototyping for AI/ML-enabled tools in support of our customer. Design and integrate LLM/agent orchestration workflows using agentic frameworks such as LangGraph, Agno, and the NVIDIA NeMo Agent Toolkit. Tune model gateway routing logic to select, evaluate, and optimize across multiple frontier and open-weight models for cost, performance, and mission fit. Develop and integrate RAG pipelines, vector databases, and memory services to give agents access to mission knowledge and context. Implement MCP (Model Context Protocol)-based tool-broker integrations so agents can safely and securely access enterprise tools and data. Evaluate, benchmark, and tune models and agent behaviors; build automated evaluation pipelines to track quality, safety, and performance over time. Partner with data scientists and mission analysts to translate research and prototypes into production-grade AI capabilities for the command. Implement AI governance guardrails — content/safety filters, policy enforcement, and approval gates — for agentic workflows, and document model/agent architectures and evaluation results for reuse. Research and evaluate emerging LLM, agent, and AI tooling to keep the command's AI capability current and adaptable. Basic Qualifications Bachelor’s degree from an accredited college or university in Computer Science, Data Science, or a related discipline and 10+ years of experience; 4 additional years of experience may be considered in lieu of degree Experience supporting mission environments Experience with frameworks such as LangGraph, Agno, or the NVIDIA NeMo Agent Toolkit Experience with model inference/serving platforms (NVIDIA NIM, vLLM, Ollama, LiteLLM) Experience with MCP (Model Context Protocol) or similar tool-broker/integration standards Experience with AI evaluation frameworks, model benchmarking, and observability (logs/metrics/traces) for AI workloads Active TS/SCI w/ polygraph Our Core Values Employees of Agile Defense are our number one priority, and the importance we place on our culture here is fundamental. Our culture is alive and evolving, but it always stays true to its roots. Here, you are valued as a family member, and we believe that we can accomplish great things together. Agile Defense has been highly successful in the past few years due to our employees and the culture we create together. What makes us Agile? We call it the 6Hs, the values that define our culture and guide everything we do. Together, these values infuse vibrancy, integrity, and a tireless work ethic into advancing the most important national security and critical civilian missions. It's how we show up every day. It's who we are. Happy - Be Infectious. Happiness multiplies and creates a positive and connected environment where motivation and satisfaction have an outsized effect on everything we do. Helpful - Be Supportive. Being helpful is the foundation of teamwork, resulting in a supportive atmosphere where collaboration flourishes, and collective success is celebrated. Honest - Be Trustworthy. Honesty serves as our compass, ensuring transparent communication and ethical conduct, essential to who we are and the complex domains we support. Humble - Be Grounded. Success is not achieved alone, humility ensures a culture of mutual respect, encouraging open communication, and a willingness to learn from one another and take on any task. Hungry - Be Eager. Our hunger for excellence drives an insatiable appetite for innovation and continuous improvement, propelling us forward in the face of new and unprecedented challenges. Hustle - Be Driven. Hustle is reflected in our relentless work ethic, where we are each committed to going above and beyond to advance the mission and achieve success. Equal Opportunity Employer/Protected Veterans/Individuals with Disabilities

Requirements

Must-haves

Tune model gateway routing logic to select, evaluate, and optimize across multiple frontier and open-weight models for cost, performance, and mission fit.
Develop and integrate RAG pipelines, vector databases, and memory services to give agents access to mission knowledge and context.
Implement MCP (Model Context Protocol)-based tool-broker integrations so agents can safely and securely access enterprise tools and data.
Evaluate, benchmark, and tune models and agent behaviors; build automated evaluation pipelines to track quality, safety, and performance over time.
Partner with data scientists and mission analysts to translate research and prototypes into production-grade AI capabilities for the command.
Implement AI governance guardrails — content/safety filters, policy enforcement, and approval gates — for agentic workflows, and document model/agent architectures and evaluation results for reuse.
Research and evaluate emerging LLM, agent, and AI tooling to keep the command's AI capability current and adaptable.
Bachelor’s degree from an accredited college or university in Computer Science, Data Science, or a related discipline and 10+ years of experience; 4 additional years of experience may be considered in lieu of degree
Experience supporting mission environments
Experience with frameworks such as LangGraph, Agno, or the NVIDIA NeMo Agent Toolkit
Experience with model inference/serving platforms (NVIDIA NIM, vLLM, Ollama, LiteLLM)
Experience with MCP (Model Context Protocol) or similar tool-broker/integration standards
Experience with AI evaluation frameworks, model benchmarking, and observability (logs/metrics/traces) for AI workloads
Active TS/SCI w/ polygraph
Our Core Values
Happy - Be Infectious. Happiness multiplies and creates a positive and connected environment where motivation and satisfaction have an outsized effect on everything we do.
Helpful - Be Supportive. Being helpful is the foundation of teamwork, resulting in a supportive atmosphere where collaboration flourishes, and collective success is celebrated.
Honest - Be Trustworthy. Honesty serves as our compass, ensuring transparent communication and ethical conduct, essential to who we are and the complex domains we support.
Humble - Be Grounded. Success is not achieved alone, humility ensures a culture of mutual respect, encouraging open communication, and a willingness to learn from one another and take on any task.
Hungry - Be Eager. Our hunger for excellence drives an insatiable appetite for innovation and continuous improvement, propelling us forward in the face of new and unprecedented challenges.
Hustle - Be Driven. Hustle is reflected in our relentless work ethic, where we are each committed to going above and beyond to advance the mission and achieve success.
Equal Opportunity Employer/Protected Veterans/Individuals with Disabilities

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

agileaimlllm

Platform: lever
Effort: quick

Benefits

Benefits detected in the listing.

Vision