AI Architect

EXL•Pittsburgh, PA

6d•$140,000 - $160,000•Remote

About The Position

We are seeking an experienced AI Architect with a critical, hard-to-source skill set in Semantic Kernel and Microsoft Agent Framework. This role involves refactoring SK plugins, planners, function calling, and memory to Agent Framework agents/workflows. You will also be responsible for rebuilding the orchestration layer using Python/Django without breaking existing API contracts, RBAC, and permissions. Experience with Azure AI Services/RAG, including Azure AI Search, Azure OpenAI, and managing retrieval, reranking, prompt assembly, and citation paths, is essential.

Requirements

Hands-on experience with Semantic Kernel + Microsoft Agent Framework, including refactoring SK plugins, planners, function calling, and memory to Agent Framework agents/workflows.
Proficiency in Python/Django for rebuilding orchestration layers while maintaining API contracts, RBAC, and permissions.
Experience with Azure AI Services/RAG, specifically Azure AI Search (hybrid retrieval, semantic ranking), Azure OpenAI (inference + embeddings), preserving retrieval, reranking, prompt assembly, and citation paths.
Graduate in Computer Science.
8+ years of experience in data engineering or a related field.

Nice To Haves

Model Context Protocol (MCP) for tool/connector integration and the Intelligence Layer channel.
OpenShift / containerization for deploying migrated microservices.
Observability & Evaluation tools such as Arize Phoenix + OpenTelemetry instrumentation.
Experience with agent/agentic-system registration.
RAG quality/eval validation to confirm parity.
GitHub / CI-CD / DevSecOps, including PR-based workflows and vulnerability remediation.
Redis Enterprise for caching/session state.
Azure APIM for gateway, rate limiting.

Responsibilities

Refactor Semantic Kernel plugins, planners, function calling, and memory to Microsoft Agent Framework agents/workflows.
Rebuild the orchestration layer using Python/Django, ensuring existing API contracts, RBAC, and permissions are maintained.
Implement and manage Azure AI Services/RAG, including Azure AI Search (hybrid retrieval, semantic ranking), Azure OpenAI (inference + embeddings), retrieval preservation, reranking, prompt assembly, and citation paths.
Integrate tools/connectors and manage the Intelligence Layer channel using Model Context Protocol (MCP).
Deploy migrated microservices on OpenShift using containerization.
Implement observability and evaluation using Arize Phoenix and OpenTelemetry instrumentation, including agent/agentic-system registration and RAG quality/eval validation.
Utilize GitHub for CI/CD and DevSecOps, including PR-based workflows and vulnerability remediation.
Implement caching and session state using Redis Enterprise.
Configure Azure APIM for gateway and rate limiting.