Agentic AI Engineer / Architect

About the role

Software Guidance & Assistance, Inc. (SGA) is searching for an Agentic AI Engineer / Architect for a contract assignment with one of our premier Human Resources clients. This role will be remote.

Responsibilities

• Assist with the creation of and support of AI capabilities in support of the Cyber Security Organization. The focus will be to enable our existing Cyber Security organization to increase capabilities, empower CSO employees, and simplify workflows. You will be instrumental in the agent framework that supports the security organization.
• Design, develop, and deploy intelligent agent systems that can reason, plan, and autonomously interact within complex environments.
• Design, build, and optimize autonomous or semi-autonomous AI agents powered by large language models (LLMs), retrieval-augmented generation (RAG), and tool-using frameworks.
• Architect systems for agent planning, memory, reflection, and recursive reasoning across multiple contexts or sessions.
• Evaluate and integrate frameworks such as LangGraph, CrewAI, AutoGen, or ReAct-style agents based on project needs.
• Implement secure and efficient agent-environment interactions (e.g., API calling, database access, toolchains).
• Build internal agent tooling to support decision-making, summarization, research, and task execution workflows.
• Collaborate with ML scientists to fine-tune or distill models for specialized agent use cases.
• Work closely with product and infrastructure teams to productionize and scale agent-based services.
• Contribute to internal best practices, safety and alignment guardrails, and monitoring of agent behaviors.
• Conduct code reviews, mentor team members, and provide strategic input into the roadmap.

Required Skills

• 6+ years in software engineering, with at least 2+ years focused on AI/ML systems or intelligent agent design.
• Proficiency with Python and experience working with AI/ML libraries such as LangChain, Hugging Face, PyTorch, Transformers, OpenAI APIs, etc.
• Strong understanding of modern LLMs, prompt engineering, and reasoning patterns (e.g., CoT, ToT, ReAct).
• Experience building complex workflows involving memory, state, external tools, and human-in-the-loop control.
• Familiarity with scalable data pipelines and backend orchestration (e.g., FastAPI, Airflow, Ray, or Kubernetes).
• Experience with vector databases (e.g., Pinecone, FAISS, Weaviate) and RAG architectures.
• Deep understanding of AI safety principles, hallucination mitigation strategies, and system monitoring.

Preferred Skills

• Experience with autonomous agents, multi-agent coordination, or simulation environments.
• Prior experience integrating with enterprise APIs, secure environments, or sensitive data contexts.
• Background in cognitive architectures, human-AI interaction, or real-time decision-making systems.
• Contributions to open-source AI agent frameworks or AI research papers.
• Experience leading cross-functional technical projects or mentoring engineering teams.