Agentic AI Engineer – Workflow Orchestration & Prompt Engineer

Experis is seeking an Agentic AI Engineer to design and improve multi-step LLM workflows that power AI agents in real execution environments. In this onsite role in Charlotte, NC, you will connect business intent to structured AI workflow architecture, combining prompt and context engineering with orchestration and operational troubleshooting.

You will translate ambiguous business challenges into reliable, staged AI solutions, then continuously evaluate and refine performance across prompts, agents, integrations, and the surrounding execution stack.

What you will do

• Design, develop, and optimize multi-step prompt workflows that support reasoning, decision support, content generation, analysis, and task execution through AI agents.
• Convert business problems into structured AI workflow designs that decompose complex work into logical stages and execution paths.
• Create and maintain reusable prompt libraries, workflow templates, and orchestration patterns, including standardized templates, version control, evaluation criteria, change tracking, and lifecycle management.
• Define workflow architecture, including prompts, context requirements, tool interactions, orchestration logic, handoffs, dependencies, and output requirements.
• Specify how information, memory, context, and intermediate outputs are managed throughout workflow execution.
• Review workflow execution logs, model outputs, and intermediate agent responses to pinpoint root causes of failures, performance degradation, or inconsistent results.
• Troubleshoot issues across prompt design, context management, agent behavior, tool integrations, data pipelines, file structures, APIs, and execution environments.
• Evaluate workflow effectiveness using qualitative and quantitative measures, then implement improvements to increase reliability, accuracy, efficiency, and user outcomes.
• Enhance context engineering approaches to improve information retrieval, grounding, memory management, and response quality.
• Build automation assets such as scripts and utilities to support workflow development, experimentation, testing, evaluation, monitoring, and operational efficiency.
• Perform data inspection, transformation, validation, and analysis to support workflow development and troubleshooting.
• Support AI governance, risk management, quality assurance, and responsible AI practices by ensuring workflows operate within established standards and controls.
• Keep up with emerging AI technologies including agent frameworks, orchestration platforms, and prompt engineering techniques.

Required qualifications

• Experience designing, implementing, and optimizing multi-step LLM, generative AI, and agentic workflows.
• Strong problem-solving skills with the ability to translate complex business challenges into scalable AI solutions.
• Expertise in prompt engineering, human-in-the-loop workflows, and AI performance optimization.
• Ability to troubleshoot and diagnose issues across prompts, agents, context management, integrations, and workflow execution.
• Proficiency in Python, APIs, JSON, and Markdown, with experience working with structured and unstructured data.
• Experience using AI-assisted development tools such as GitHub Copilot, Claude Code, Cursor, or similar platforms.
• Knowledge of agent orchestration frameworks, multi-agent architectures, and modern workflow orchestration patterns.
• Understanding of RAG, context engineering, memory management, tool calling, and agent planning methodologies.
• Experience integrating AI solutions with enterprise applications, automation platforms, cloud AI services, and business processes.
• Strong skills in communication, collaboration, governance, risk management, and responsible AI, including evaluation, monitoring, compliance, and operational controls.

Technologies and focus areas

• Python, APIs, JSON, Markdown
• GitHub Copilot, Claude Code, Cursor
• RAG, tool calling, and multi-agent architectures
• Agentic AI and workflow orchestration: tool calling, handoffs, orchestration patterns
• Prompt and context engineering: RAG, grounding, context/memory management, human-in-the-loop patterns
• LLM evaluation and troubleshooting: testing workflows, analyzing outputs and logs, improving accuracy and reliability
• Business-to-AI solution design: translating ambiguous needs into scalable, structured end-to-end AI solutions