Agentic AI Engineer (Only Visa Independent Candidate/W2)

Responsibilities

• Design, build, and deploy AI agents that retrieve information, call tools, and execute multi-step workflows with human review where needed.
• Build the orchestration layer that lets agents execute reusable skills, manage context and state, recover from errors, and track execution.
• Integrate agents with enterprise APIs, document repositories, databases, and governance, risk, and compliance systems.
• Work with audit, risk, and business stakeholders to automate evidence collection, document review, population and sample analysis, control validation, and exception identification.
• Make results traceable to supporting evidence. Implement evaluations, logging, monitoring, access controls, and review checkpoints. Measure accuracy, consistency, cost, and performance against defined benchmarks.

Required experience

• Hands-on experience building and deploying agentic applications that execute multi-step workflows.
• Strong Python and/or TypeScript skills, including backend development and API integration.
• Experience implementing tool calling, structured outputs, workflow orchestration, agent state management, and reusable agent capabilities.
• Experience grounding and evaluating LLM outputs, handling exceptions, and securing applications that interact with enterprise data.
• Ability to translate business requirements into working software and communicate with technical and business stakeholders.

Preferred experience

• Direct experience building AI harnesses that load and execute reusable skills.
• Applying agentic AI to internal audit, SOX/IT testing, compliance assessments, or other controls testing.
• Familiarity with control design, operating effectiveness, evidence sufficiency, sampling, and exception documentation.
• Experience integrating with GRC, identity, ticketing, or ERP systems and implementing scoped permissions, approval gates, sandboxed execution, or auditable agent activity.