Generative AI Engineer

Centraprise; Charlotte (NC); On-site; USD 140,000 - 190,000; Full time

Job summary

Centraprise is seeking a hands-on Lead Agentic AI & GenAI Engineer to architect and productionize secure, autonomous and human-supervised AI solutions. The role emphasizes agent design, LLM application engineering, AI-assisted software delivery, evaluation, safety, and production operations, with strong data integration skills to connect governed enterprise sources.

The ideal candidate will design and implement robust agent workflows, integrate with enterprise APIs and databases, and ensure

Qualifications

• Experience designing and deploying AI agents in enterprise environments.
• Strong data integration skills to connect governed enterprise sources.

Responsibilities

• Architect single-agent and multi-agent solutions using AI tooling and frameworks.
• Design agent workflows with context protocols, function calling, and memory.
• Implement agent services with Python/TypeScript, asynchronous execution, and robust fault handling.
• Build grounded enterprise AI experiences with retrieval, knowledge graphs, and governance.
• Evaluate agents with benchmarks, tracing, latency controls, and production SLOs.
• Ensure AI security including prompt defense, least-privilege access, and auditability.
• Collaborate with product, operations, risk, cyber, and platform teams to integrate with systems of record.

Job description

Overview:

We are seeking a hands-on Lead Agentic AI & GenAI Engineer to architect and productionize secure, autonomous and human-supervised AI solutions. The role emphasizes agent design, LLM application engineering, AI-assisted software delivery, evaluation, safety, and production operations, with strong data integration skills to connect governed enterprise sources.

Responsibilities

• Architect single-agent and multi-agent solutions using Google ADK, LangGraph/LangChain, Microsoft Semantic Kernel, AutoGen, OpenAI Agents SDK.
• Design agent workflows using Model Context Protocol (MCP), function calling, structured outputs, reusable skills, planning, state, memory, A2A handoffs, and human approval gates.
• Use Claude Code, GitHub Copilot agent mode/coding agent, Devin, Cursor, OpenAI Codex for implementation, testing, refactoring, documentation, and pull request workflows.
• Build grounded enterprise AI experiences using advanced RAG, hybrid retrieval, reranking, knowledge graphs, citations, context engineering, and governed workflows.
• Engineer reliable Python and TypeScript agent services with asynchronous execution, retries, checkpointing, idempotency, rollback, escalation, and long-running workflow patterns.
• Implement agent evaluation and LLMOps using task-success benchmarks, ARIZE tracing, prompt/version management, latency and token-cost controls, and production SLOs.
• Implement enterprise AI security including prompt-injection defense, least-privilege tool access, sandboxing, PII protection, content safety, auditability, and red-team, and policy enforcement.
• Integrate agents with enterprise APIs, databases, events, documents, and Systems of Record, partnering with product, operations, risk, cyber, architecture, and platform teams.