On-Site AI Engineer - Secure, Production Agent Architect

acuCyber; Reston (VA); On-site; USD 120,000 - 190,000; Full time

Job summary

acuCyber seeks an AI Engineer to design and deploy secure AI systems connecting foundation models with enterprise tools, data, and services. You will build agent architectures, advanced retrieval capabilities, and stateful workflows that perform complex tasks while enforcing user permissions and maintaining reliable, controlled execution.

We are looking for a hands-on engineer who understands autonomous agents and can operate them in production with a focus on security, observability, and human

Responsibilities

• Agentic Architecture & Protocol Integration: Architect modular, interoperable agent ecosystems utilizing the Model Context Protocol (MCP) to standardize connections between foundation models, external tools, and enterprise data services. Design complex stateful workflows using production orchestration frameworks (LangGraph, CrewAI, Claude Agent SDK).
• Next-Gen RAG & Stateful Reasoning: Engineer advanced retrieval systems including Graph-based RAG, Agentic Routing, and Corrective RAG (CRAG). Design memory architectures that optimize context window utilization and reliably manage state across long-lived, multi-turn agent sessions.
• Data Governance & Entitlement Enforcement: Establish strict permission boundaries around autonomous tool calling. Map strict data serialization schemas (JSON Schema, Protocol Buffers) directly to enterprise Access Control Lists (ACLs) and RBAC/ABAC policies so agents only retrieve data and execute actions explicitly authorized for the invoking end-user.
• Runtime Guardrails & System Stability: Build resilient execution harnesses that prevent runaway agent loops and cascading failures. Implement idempotent tool retries, strict token and cost budgets, deterministic early-exit stop conditions, and seamless Human-in-the-Loop (HITL) approval gates for high-consequence actions

Job description

acuCyber seeks an AI Engineer to design and deploy secure AI systems connecting foundation models with enterprise tools, data, and services. You will build agent architectures, advanced retrieval capabilities, and stateful workflows that perform complex tasks while enforcing user permissions and maintaining reliable, controlled execution.

We are looking for a hands-on engineer who understands autonomous agents and can operate them in production with a focus on security, observability, and human