Agentic AI Engineer

Booz Allen Hamilton•Washington, DC

1d•Remote

About The Position

Agentic AI Engineer The Opportunity: We are looking for a highly skilled Agentic AI Engineer to join our team, specializing in building autonomous, goal-oriented AI systems. You will play a crucial role in shifting our AI strategy from passive LLM chatbots to proactive, multi-agent orchestrations. In this role, you will utilize deep experience in agent orchestration frameworks, RAG, knowledge graphs, and fine-tuning Small Language Models (SLMs) for edge deployment.

Requirements

5+ years of experience in software development
3+ years of experience with GenAI, LLMs, and agentic workflows
3+ years of experience with LangChain, LangGraph, AutoGen, or LlamaIndex
Experience with MCP for tool integration and A2A for agent-to-agent collaboration
Experience with RAG architecture and KG, including Neo4j or NebulaGraph
Experience fine-tuning LLMs or SLMs using Hugging Face, PEFT, or LoRA
Knowledge of modern software design patterns, including microservice design or edge computing
Ability to obtain a Secret clearance
Bachelor’s degree in a CS or AI field

Nice To Haves

Experience deploying agentic systems in a production environment
Experience deploying agents on edge devices such as Android or local models
Experience integrating coding agents such as Cursor or Windsurf, into an efficient development pipeline with measured results
Master’s degree in a CS or AI field preferred; Doctorate degree in CS or Statistics a plus

Responsibilities

Design and implement intelligent agent architectures that can reason, plan, and take actions using LangChain, LangGraph, and AutoGen.
Develop and deploy multi-agent systems using Model Context Protocol (MCP) and Agent-to-Agent (A2A) protocols to facilitate communication, tool usage, and collaborative task solving.
Build advanced RAG pipelines integrating unstructured data with Knowledge Graphs (KG) to enhance reasoning accuracy and context retention.
Fine-tune SLMs for specific domains and optimize them for edge device performance, including ONNX, GGML, or Ollama.
Develop evaluation frameworks to test agent reliability, safety, and performance, moving from prototype to production, including ReAct loops and human-in-the-loop.