AI Engineer Lead

Role Summary

We are seeking a Senior AI Engineer specialized in Microsoft Azure to lead the design, development, and deployment of agentic AI systems and enterprise-grade GenAI platforms. You will serve as a technical leader—defining architecture patterns (including Model Context Protocol – MCP), building RAG and LLMOps foundations, and ensuring solutions meet financial-services-grade security, compliance, and reliability standards.

Key Responsibilities

Agentic AI Architecture & Development

• Design and implement agentic AI solutions (autonomous workflows, multi-agent systems, tool-use) using Azure OpenAI / Azure AI Studio and Semantic Kernel/Lang Chain/Lang Graph.

• Define and enforce architecture patterns for agents, tools, memory, and MCP integrations.

• Implement schema-constrained prompting, output validation, and guardrails to reduce hallucinations and enforce safety.

GenAI Pipelines & RAG

• Build modular pipelines for ingestion, transformation, and orchestration across LLMs, embedding services, vector stores, SQL sources, and APIs.

• Develop hybrid RAG using Azure AI Search (vector & hybrid) plus structured sources (e.g., Azure SQL/SQL MI, Synapse, APIs).

• Implement post-retrieval re-ranking, chunking strategies, feedback/DPO-based ranking, and caching (KV) for latency & cost control.

• Monitor performance and quality with Langfuse/Prompt Flow traces, Azure Monitor, and Prometheus/Grafana.

AI Platform & Services

• Architect platform components: embedding pipelines, vector stores (Azure AI Search vectors, Cosmos DB with vector support, Redis/pgvector on Azure), document intelligence (Azure Document Intelligence / OCR), and metadata extraction.

• Define API standards and integration patterns (REST/GraphQL/gRPC) backed by Azure API Management.

Cloud Architecture & Infrastructure (Azure-first)

• Lead cloud architecture on Azure for scalability, security, and cost: AKS, Azure Functions, Event Hubs (Kafka-compatible), Service Bus, Data Lake Storage Gen2, Blob Storage, Key Vault, Private Link/VNet.

• Productionize models using Azure Machine Learning (model registry, managed endpoints, Prompt Flow, MLflow integration).

• Establish infrastructure standards for containerization (Docker), orchestration (AKS / ARO-OpenShift), monitoring, and MLOps/LLMOps.

Engineering Excellence & Leadership

• Lead design reviews, code reviews, and CI/CD (GitHub Actions or Azure DevOps) with automated prompt tests, guardrail tests, and regression suites.

• Mentor AI engineers; build internal best practices and knowledge assets.

• Define SLOs/SLAs and implement proactive alerting, rate-limits, and fallbacks.

Stakeholder & Governance

• Translate business requirements into technical designs with AI Product Owners and domain stakeholders.

• Partner with Security, Risk, and Enterprise Architecture on standards, privacy, and Responsible AI practices (e.g., Azure Content Safety).

Required Skills & Qualifications

Education

• Bachelor’s/master’s in computer science, AI/ML, Data/Software Engineering, or related field.

Experience

• 10+ years in software/data/ML engineering, including hands-on production delivery of AI/ML solutions.

• 3+ years delivering GenAI/LLM and RAG systems in production (Azure preferred).

• Proven leadership of complex technical initiatives in enterprise or financial services environments.

Technical Skills (Azure-Focused)

• LLMs & GenAI: Azure OpenAI (GPT, vision, embeddings), model selection, prompt engineering, function/tool calling, LLMOps (prompt versioning, evaluation, output validation).

• Orchestration: Semantic Kernel, Lang Chain/LangGraph, tool routing, multi-hop workflows, MCP.

• RAG & Vectors: Azure AI Search (hybrid+vector), Cosmos DB vectors/Redis/pgvector, FAISS/Qdrant (self-hosted on AKS).

• Data & Ingestion: Azure Document Intelligence (Form Recognizer/OCR), Cognitive Services, Azure Data Factory/Synapse pipelines, APIs/SQL; layout-aware parsing & chunking.

• MLOps: Azure ML (registries, pipelines, endpoints), MLflow, feature/model/version management, A/B and shadow deployments.

• Cloud & Infra: AKS/ARO, Docker, Event Hubs (Kafka), Service Bus, Functions, API Management, Key Vault, Monitor, Log Analytics, Application Insights, Managed Identities, VNet/Private Link.

• Programming: Python (advanced); experience with Transformers, Sentence Transformers, PyTorch.

• Services & APIs: REST/gRPC design, auth (Entra ID/OAuth2), RBAC, multi-tenant patterns; caching and rate limiting.

• Observability: Azure Monitor, Prometheus/Grafana, Langfuse or equivalent.

Industry & Governance

• Experience in financial services with understanding of regulatory, model risk management, security, privacy, and auditability requirements.