Responsibilities
- Architect end-to-end GenAI-first and Agentic AI solutions on Azure using: Azure OpenAI Service
- Architect end-to-end GenAI-first and Agentic AI solutions on Azure using: Azure AI Foundry
- Architect end-to-end GenAI-first and Agentic AI solutions on Azure using: Azure Agents (Agent Service + Agent Runtime)
- Architect end-to-end GenAI-first and Agentic AI solutions on Azure using: Microsoft Agent Framework SDK
- Architect end-to-end GenAI-first and Agentic AI solutions on Azure using: Azure Machine Learning (Azure ML) for model lifecycle, fine-tuning & evaluation
- Design multi-agent systems capable of planning, reasoning, memory management, tool invocation, and complex workflow execution.
- Architect enterprise-grade RAG systems using: Azure Cognitive Search
- Architect enterprise-grade RAG systems using: Azure AI Search Vector Stores
- Architect enterprise-grade RAG systems using: Cosmos DB + pgvector
- Architect enterprise-grade RAG systems using: Redis Enterprise on Azure
- Build secure prompt pipelines, embeddings workflows, tool integration layers, and model evaluation frameworks optimized for Azure’s GenAI stack.
- Build Python-based microservices (FastAPI/Django) tightly integrated with Azure AI and cloud-native components.
- Implement enterprise-grade LLMOps using: Azure ML Pipelines
- Implement enterprise-grade LLMOps using: Azure DevOps (ADO)
- Implement enterprise-grade LLMOps using: GitHub Actions for MLOps/LLMOps
- Implement enterprise-grade LLMOps using: Azure Monitor, App Insights, Log Analytics for observability
- Establish engineering standards for: Agent governance & safety
- Establish engineering standards for: Content filters & guardrails (Azure OpenAI Safety Systems)
- Establish engineering standards for: Secure model invocation via private endpoints
- Establish engineering standards for: Latency reduction, performance tuning & cost governance
- Optimize AI workloads using model caching, selective batching, adaptive routing, vector index optimization, and autoscaling strategies on AKS/Functions.
- Lead architecture reviews, threat modeling, cost modeling, and performance benchmarking for large-scale GenAI workloads.
- Evaluate new Azure GenAI capabilities, FMs, agent frameworks, vector databases, and orchestration tools.
- Build internal accelerators and reusable reference architectures for: Multi-agent orchestration with Azure Agents Framework, Langgraph, CrewAI
- Build internal accelerators and reusable reference architectures for: RAG 2.0 and context enrichment pipelines
- Build internal accelerators and reusable reference architectures for: Federated & hybrid retrieval patterns
- Build internal accelerators and reusable reference architectures for: Enterprise system/tool integration
- Drive PoCs, prototypes, and tech spikes to validate emerging Azure GenAI capabilities and accelerate enterprise adoption.
Requirements
- Microsoft Certified Azure Solutions Architect Expert or equivalent Azure Architect certification.
- Strong hands-on experience designing AI/ML or GenAI systems on Azure.
