Senior Azure AI Engineer
Summary
Design and deploy AI/LLM-powered applications on Azure, building end-to-end pipelines, prompt flows, and RAG solutions while ensuring Responsible AI practices and cloud-native integration.
Job Description:
- Design, build, and deploy AI models and end-to-end AI pipelines for production environments
- Integrate AI capabilities into applications and services using APIs and cloud-native architectures
- Collaborate with Data Engineers, Software Engineers, and Product Teams to ensure seamless data flow and system integration
- Monitor, evaluate, and optimize model performance, accuracy, and scalability in real-world use
- Develop and manage Prompt Flows, orchestration pipelines, and agent-based AI workflows
- Implement Retrieval-Augmented Generation (RAG) solutions, including embedding, indexing, and context management
- Ensure adherence to Responsible AI practices, including model safety, governance, and compliance standards
- Establish observability, logging, and performance monitoring for AI systems
- Apply secure-by-design principles, including identity management, access control, and data protection
- Translate business requirements into AI solutions, defining guardrails, KPIs, and success criteria.
Requirements
- Related Work Experience - 3–6+ years in AI/ML, software engineering, or cloud-based AI solution development a. Hands-on experience building and deploying AI/LLM-powered applications in production
- b. Experience with Azure or similar cloud platforms (AWS/GCP)
- c. Proven work on prompt engineering, orchestration, or RAG-based solutions
- d. Experience collaborating in cross-functional product or engineering teams
- Knowledge – knowledgeable in the following:
- a. AI/LLM engineering: prompt design, orchestration (Prompt Flow), agent-based systems, and RAG implementation
- b. Software engineering: Python, APIs (REST/JSON), microservices, and CI/CD practices
- c. Azure AI ecosystem: AI Foundry, model deployment, inference APIs, and cost optimization
- d. Data and search: embeddings, chunking strategies, and Azure AI Search (hybrid retrieval)
- e. Cloud and security: Azure networking, identity (Entra ID), observability, and secure-by-design architectures
- f. Responsible AI: model safety, governance, explainability, and policy enforcement
- g. Low-code integration: Copilot Studio and Power Platform extensibility
MGL