Agentic AI Engineer
Job Description
Agentic AI Engineer role based in Arlington, VA onsite, focusing on designing, building, and delivering production-grade agentic AI systems and modern GenAI applications.
Responsibilities
- Create and deploy production-ready agentic AI systems that demonstrate practical value from generative AI, LLMs, and autonomous workflows.
- Design modular, reusable AI app patterns; integrate cloud-hosted, local, and hybrid model providers; leverage structured prompting, tool use, orchestration, and multimodal reasoning.
- Build solutions deployable across cloud platforms and portable, self-contained builds, optimizing latency, cost, observability, and safety.
- Prototype and iterate rapidly with AI-assisted development tools, validating ideas through evaluation-driven development and continuous experimentation.
- Define the direction of mission-critical agentic systems by selecting prompting strategies, RAG architectures, agentic workflows, and suitable models.
- Collaborate with data engineers, data scientists, solutions architects, and product owners to deliver robust, scalable solutions.
Requirements
- Minimum 2 years designing, building, or deploying AI-driven systems, including autonomous agents, LLM-based systems, or automated decision pipelines.
- At least 2 years programming in an object-oriented language such as Python for AI/ML solution development.
- Experience with agent orchestration frameworks like LangChain, AutoGen, CrewAI, or custom agent platforms.
- Experience integrating LLMs, GPT-class models, or multimodal models into applications, pipelines, or mission systems.
- Familiarity with RAG architectures, evaluation methodologies, experimentation workflows, and asynchronous or event-driven programming patterns.
- Experience in cross-functional delivery environments with data scientists, ML engineers, SREs, product managers, and security teams; creating reference architectures and technical roadmaps.
- Secret clearance.
- Bachelor's degree.
Technologies
- Python
- LangChain
- AutoGen
- CrewAI
- GPT-class models
- LLMs
- Retrieval-augmented generation (RAG)
- Multimodal models
Benefits
- Health, life, disability, financial, and retirement benefits
- Paid leave
- Professional development
- Tuition assistance
- Work-life programs
- Dependent care
- Employee recognition awards program
Nice If You Have
- Experience with agent frameworks, interoperability standards, and multi-agent patterns such as MCP, A2A, LangGraph, or equivalents
- Experience with model fine-tuning, prompt tuning, domain adaptation, or reinforcement learning from human or AI feedback
- Experience designing evaluation suites or safety testing frameworks for AI systems, and integrating AI with external tools, APIs, or enterprise systems via tool calling
Clearance
- Secret clearance is required; applicants must meet eligibility for access to classified information.
Compensation
- Salary range: $99,000 to $225,000 USD per year
- Compensation reflects location, education, knowledge, skills, competencies, experience, contract affordability, and organizational needs
- Posting closes within 90 days from the posting date
Identity Statement
- Identity verification involves biometrics and AI to ensure authenticity and reduce fraud
- Interviews and assessments will be on camera
- Booz Allen may capture your photo for identity verification purposes
Candidate AI Usage Policy
- AI is part of daily work, and Booz Allen advocates responsible use of AI tools
- Use of AI or tools to assist with interview responses is prohibited unless permission is granted
Work Model
- Onsite: work primarily at a Booz Allen office or customer facility, collaborating with colleagues and clients as required
Commitment to Non-Discrimination
- All qualified applicants will receive consideration without regard to disability, protected veteran status, or any other status protected by law