Senior Agentic AI Engineer

the Client•Glendale, AZ

2d•Hybrid

About The Position

We are seeking a hands-on Senior Agentic AI Engineer to design, build, and operate production-grade Agentic AI, Decision Intelligence, and Retrieval-Augmented Generation solutions on Microsoft Azure for a leading engineering organization's Digital Supply Chain Systems organization. The role will support diverse and evolving DSCS business requirements by developing reusable, scalable, secure, and explainable enterprise AI capabilities.

Requirements

Strong programming proficiency in Python
working knowledge of TypeScript/Node.js or a comparable language for application services.
Strong software engineering background with experience designing and deploying production-grade cloud applications.
Hands-on experience building Generative AI and RAG applications with Azure AI Foundry, Azure OpenAI, Azure AI Search, LLM APIs, embeddings, vector or hybrid search, knowledge retrieval, grounding, and citations.
Experience with Agentic AI frameworks such as Microsoft Agent Framework, Semantic Kernel, LangGraph, AutoGen, or comparable frameworks, including single-agent and multi-agent systems, tool calling, MCP-based tool integration, and human-in-the-loop controls.
Experience building natural language to SQL or conversational analytics solutions, including schema and metadata modeling, query generation, query validation, and grounding answers in retrieved data.
Experience designing or contributing to decision intelligence systems that combine AI, analytics, business context, and operational workflows to improve decision quality and business outcomes.
Experience evaluating and improving agent quality through prompt engineering, test datasets, LLM-based evaluation, safety checks, reasoning-quality assessment, and production feedback loops.
Strong knowledge of LLMOps , CI/CD, Docker and Kubernetes, observability, and production operations for AI applications.
Working knowledge of core Azure platform services: AKS or Azure Container Apps, Azure Functions, API Management, Entra ID, Key Vault, and Azure SQL or Cosmos DB.
Good understanding of RESTful APIs, asynchronous patterns, secure integrations, relational databases, SQL, SQL/NoSQL data stores, and data engineering or ETL pipelines.
Experience with enterprise-scale secure AI deployments, including identity, authorization, data privacy, compliance, and production monitoring.
Strong analytical, problem-solving, collaboration, and communication skills.
8 years of relevant software engineering experience.
3 years of hands-on experience developing Generative AI, machine learning, intelligent automation, or LLM-based applications, including substantial recent experience with Agentic AI.

Nice To Haves

Experience with Microsoft Fabric, Azure Databricks, or Azure Data Factory for AI-ready data pipelines.
Exposure to Copilot Studio, Power Platform, or Teams-based agent experiences.
Experience with model fine-tuning (e.g., LoRA/QLoRA), prompt caching, and token/cost optimization at scale.
Supply chain domain knowledge (procurement, expediting, logistics, materials management) or familiarity with ERP data such as Oracle EBS or SAP.
Front-end experience with React and TypeScript for building agent-facing user interfaces.
Microsoft certifications such as Azure AI Engineer Associate (AI-102/AI-103) or Azure Solutions Architect (AZ-305).

Responsibilities

Design and build production-grade single-agent and multi-agent systems in Azure AI Foundry capable of orchestrating reasoning, planning, tool usage, and workflow execution.
Architect scalable Generative AI solutions leveraging Azure AI Foundry, Azure OpenAI, Azure AI Search, and other enterprise AI services.
Build and govern Retrieval-Augmented Generation (RAG) architectures using structured and unstructured enterprise data sources, including embeddings, vector/hybrid search, chunking, metadata filtering, reranking, grounding, and citations.
Develop secure, reliable integrations between agents and enterprise systems using Model Context Protocol (MCP), REST APIs, relational databases, and event-driven services.
Design Natural Language to SQL (NL2SQL) capabilities, ensuring accuracy, explainability, and secure access to enterprise data.
Define semantic-layer strategies that enable AI agents to understand enterprise data models, business metrics, terminology, and relationships.
Architect decision intelligence capabilities that combine enterprise data, business context, analytics, and AI reasoning to support informed decision-making.
Establish reusable frameworks for impact analysis, dependency identification, prioritization, recommendation generation, and decision traceability.
Ensure AI-generated recommendations are explainable, evidence-based, grounded in authoritative enterprise data, and aligned with business objectives.
Define and automate evaluation methodologies for response quality, reasoning quality, recommendation relevance, business usefulness, and user trust, including golden datasets, LLM-based evaluation, and regression evals wired into CI/CD.
Implement guardrails and Responsible AI controls: input/output content safety, PII protection, grounding checks, authorization boundaries, and human-in-the-loop escalation paths.
Establish governance, security, safety, transparency, and compliance standards for enterprise AI solutions.
Implementation of observability and monitoring frameworks for AI applications, including quality, performance, reliability, and adoption metrics, plus traces, tool calls, latency, token usage, and cost.
Optimize solutions for cost and latency through model selection, prompt and context engineering, caching, and workload right-sizing.
Drive architectural decisions for scalable, cloud-native AI platforms using modern software engineering, DevOps, and MLOps/LLMOps practices.
Define technical standards, reference architecture, and reusable frameworks for AI agents, reasoning systems, and decision-support applications.
Mentor engineers and provide technical framework on Agentic AI, decision intelligence, software architecture, and enterprise AI best practices.
Partner with business stakeholders and domain experts to transform complex business requirements into reusable AI capabilities.
Collaborate with data engineering teams to develop AI-ready data products, semantic models, metadata frameworks, and enterprise knowledge layers.
Stay current with emerging AI technologies, frameworks, and industry practices, evaluating their applicability within DSCS and enterprise environments.