Remote Agentic AI Architect

Booz Allen Hamilton

Washington (Washington County)

Hybrid

USD 135,000 - 225,000

Full time

14 days+

Get more replies from employers

Send a job-specific resume in minutes.

Benefits offered by this job

TS/SCI clearance with polygraph

Competitive benefits package

Job summary

Booz Allen Hamilton seeks an experienced Agentic AI Engineer to shift AI strategy toward proactive multi-agent orchestration. The role focuses on agent architectures, MCP/A2A protocols, and edge-optimized LLM fine-tuning using SLMs and knowledge graphs.

Candidates should have 4+ years in software, 3+ years in production ML, cloud experience (AWS/Azure), and TS/SCI clearance. The position offers a collaborative, fast-paced environment with cutting-edge AI tooling.

Qualifications

4+ years of software development experience.
3+ years building production ML solutions with Docker/Kubernetes.
3+ years with LangChain, LangGraph, AutoGen, PydanticAI, CrewAI, or LlamaIndex.
3+ years cloud experience (AWS and Azure).
Experience with MCP for tool integration and A2A collaboration.
Experience with RAG architectures and Knowledge Graphs (Neo4j/NebulaGraph).
Fine-tuning LLMs/SLMs with Hugging Face, PEFT, or LoRA.
Observation layers with Grafana, Langfuse, LangSmith, or Phoenix.
Knowledge of microservice design or edge computing.
TS/SCI clearance with polygraph; willingness to undergo.

Responsibilities

Design and implement agent architectures that can reason, plan, and act.
Develop multi-agent systems using MCP and A2A protocols for communication and tool usage.
Build RAG pipelines integrating unstructured data with knowledge graphs to improve context.
Fine-tune SLMs for edge deployment (ONNX/GGML/Ollama) and optimize performance.
Develop evaluation frameworks for reliability, safety, and production readiness; include ReAct loops.