Key Responsibilities

1. AI & Fabric Data Agent Development

• Build & Deploy Data Agents: Design, test, and maintain Fabric-native Data Agents using Azure OpenAI Assistant APIs to enable natural language querying across Lakehouse, Warehouses, and KQL databases.
• Agentic Prompt Engineering: Design system prompts, few-shot examples, domain-specific instructions, and chain-of-thought instructions to ensure AI agents generate accurate, secure, and context-aware SQL/DAX queries.
• Guardrails & Governance: Integrate Azure AI Content Safety and Microsoft Purview policies to enforce read-only data access, security boundaries, and prevent hallucinated or unauthorized outputs.

2. Microsoft Fabric & Semantic Modelling

• AI-Ready Architecture: Clean and structure data in Fabric OneLake to be "AI-ready". Optimize tables, columns, and relationship naming conventions so LLMs can interpret schemas accurately.
• Semantic Model Design: Develop highly optimized Power BI Semantic Models with Direct Lake connectivity to provide the single source of truth for both human-facing dashboards and AI agents.
• Data Orchestration: Build and troubleshoot pipelines using Dataflows Gen2 and Data Factory inside Fabric to ingest diverse enterprise data.

3. BI & Dashboards

• Interactive Visualizations: Design and build executive-level Power BI dashboards with intuitive UX, focusing on actionable insights.
• Hybrid Delivery: Ensure that users can transition seamlessly between structured reports (Power BI) and conversational analytics (Fabric Data Agents).
• Advanced Analytics: Implement complex DAX measures, calculations, and analytical patterns to support diagnostic and predictive business questions.

Required Skills & Experience

Technical Stack

• Platform: 2+ years of hands-on experience with Microsoft Fabric (OneLake, Lakehouse, Warehouses, and Real-Time Intelligence).
• BI & Analytics: Strong proficiency in Power BI, advanced DAX, and semantic data modelling.
• AI & Prompt Engineering: Solid understanding of Large Language Models (LLMs), prompt engineering techniques , and RAG-based architectures.
• Data Languages: Strong SQL and Python skills for data preparation and pipeline scripting.