Must Have Technical/Functional Skills
- Agentic AI: AI agents, planning, reasoning, memory, tool/function calling, multi-agent orchestration, guardrails, and agent evaluation.
- Frameworks: LangChain, LangGraph, Semantic Kernel, AutoGen, CrewAI, or equivalent agent orchestration frameworks.
- LLM / GenAI: Azure OpenAI/OpenAI, RAG, embeddings, vector search, hybrid search, prompt engineering, grounding, and hallucination mitigation.
- Microsoft Fabric: Lakehouse, Warehouse, OneLake, Data Factory, Dataflows, Notebooks, Power BI semantic models, Direct Lake, Delta Lake, and medallion architecture.
- Ontology / Semantic Modeling: business entities, relationships, hierarchies, taxonomy, metadata, lineage, semantic layer, knowledge graphs, RDF/OWL/SPARQL or graph databases.
- Engineering: Python, SQL, Spark/PySpark, REST APIs, CI/CD, Git, monitoring, logging, cloud security, and production support.
- Functional: ability to translate business processes into ontology models, semantic data products, and agent workflows
Roles & Responsibilities
- Design and implement enterprise-grade agentic AI solutions that retrieve information, call tools/APIs, reason over data, and automate workflows.
- Integrate AI agents with Microsoft Fabric Lakehouse/Warehouse, semantic models, APIs, enterprise systems, and document repositories.
- Build and optimize RAG pipelines using embeddings, vector/semantic search, structured data grounding, and evaluation datasets.
- Design ontology-driven semantic models covering entities, relationships, metadata, business rules, lineage, and governance.
- Develop Fabric data pipelines using Data Factory, Notebooks, Spark/PySpark, SQL, Delta Lake, and medallion architecture.
- Implement AI guardrails, access control, logging, monitoring, cost optimization, and responsible AI practices.
- Create technical designs, architecture diagrams, reusable components, coding standards etc.