Senior AI Systems & Applications Engineer

SUMMARY:

We are seeking a versatile AI Solutions Architect to lead the design and implementation of our next-generation agentic AI ecosystem. You will bridge the gap between data engineering and application development, leveraging the full Microsoft Azure AI Foundry (formerly AI Studio) and MS Fabric suite.

The ideal candidate doesn't just "call an API"; they architect secure, governed, and multi-agentic systems that integrate deeply with enterprise data via OneLake and stay protected through MS Purview. You will be responsible for building the "connective tissue" of our AI strategy—from MCP server integration to complex prompt orchestration.   

DUTIES AND RESPONSIBILITIES: 

• AI Orchestration: Design and deploy agentic and multi-agentic workflows using Azure AI Foundry, Semantic Kernel, or AutoGen. 

• Data Strategy: Architect and manage data pipelines within MS Fabric OneLake, ensuring "Gold" layer data is ready for RAG (Retrieval-Augmented Generation) and model fine-tuning. 

• Security & Identity: Execute application and agent registration via MS Entra (formerly Azure AD), implementing OAuth flows and Managed Identities for secure tool access. 

• Governance: Configure and maintain MS Purview to ensure data lineage, sensitivity labeling, and AI safety guardrails are enforced across all LLM interactions. 

• Connectivity: Build and maintain MCP (Model Context Protocol) servers and custom tools to allow AI agents to interact securely with internal databases and third-party SaaS. 

• Development: Build production-ready application interfaces (React/Next.js or Python-based frameworks) that integrate seamlessly with backend AI flows. 

REQUIRED TECHNICAL SKILLS:  

Microsoft AI & Data Stack 

• Azure AI Foundry / MS Foundry: Deep expertise in Prompt Flow, model evaluation, and deployment of GPT-4o/o1 and SLMs (Phi-3). 

• MS Fabric & OneLake: Experience with Lakehouses, Shortcuts, and integrating Fabric data with AI search indexes. 

• MS Purview: Implementation of data loss prevention (DLP) and sensitivity labels within AI-generated content. 

• MS Entra: Advanced knowledge of App Registrations, Service Principals, and Scopes for AI Agents. 

 Generative & Agentic AI 

• Prompt Engineering: Advanced techniques (Chain-of-Thought, ReAct, Few-shot) and prompt versioning. 

• Agentic Frameworks: Proficiency in building autonomous agents that can use tools, manage state, and collaborate (multi-agentic patterns). 

• MCP (Model Context Protocol): Experience implementing MCP servers to standardize how LLMs access local and remote data. 

Software Engineering