Manager AI Software Engineering

About the role

Dubai Holding is hiring a Manager AI Software Engineering for its corporate office, reporting to the Associate Director Software Engineering. The role designs, builds, deploys and operates production-grade AI agents on Microsoft Foundry, a hands-on, pro-code position where the engineer writes Python, integrates Azure services and ships agentic applications. It works in agile teams with Product, Data Science and Architecture, and focuses on technical delivery while product priorities are set by Product Managers.

Key responsibilities

• Design and build AI agents with Microsoft Foundry, using its agent framework, model catalogue and tool/function-calling capabilities
• Implement AI solutions in Python with patterns such as RAG, tool use, structured outputs, memory and multi-agent orchestration
• Integrate agents with Palantir Foundry via MCP to access enterprise data
• Deploy agents in production using CI/CD pipelines with automated evaluations as release gates, and manage versioning of code, prompts and configurations
• Optimize agent architectures and prompts for performance, cost and latency, and handle observability, tracing, cost monitoring, regression testing and incident response
• Maintain secure data handling with Azure Key Vault and Entra ID
• Integrate applications with Azure OpenAI, AI Search, Storage, Functions, Container Apps and identity services, and contribute reusable components and shared libraries
• Build evaluation harnesses and guardrails for groundedness, safety, latency and cost, and iterate on prompts and architectures
• Work in agile squads with Product Managers, Data Scientists and Architects, communicating technical trade-offs to stakeholders

Required profile

• 2+ years of experience deploying agentic AI systems
• Bachelor's degree in computer science, engineering or a related quantitative field; advanced degrees preferred
• Hands-on production experience building LLM-based or agentic systems (tool use, orchestration, RAG, evals)
• Strong ownership, structured problem solving, clear communication and adaptability in fast-paced environments

Required skills

• Strong Python, idiomatic, typed and tested, with async and modern packaging
• Production AI agent frameworks such as Microsoft Foundry, Semantic Kernel, AutoGen, LangGraph, LlamaIndex or similar
• Azure exposure (OpenAI, AI Search, identity, storage, serverless) and Azure DevOps or GitHub Actions
• LLM patterns: tool/function calling, structured outputs, RAG, memory, multi-agent orchestration
• Git, CI/CD, containers and infrastructure-as-code; data engineering and vector stores a plus