Overview
Skills
Job Details
AI Agent Architecture
Design agents connecting primarily to Microsoft Fabric Data Lake as central data repository
Access structured data from Infor M3 ERP, Salesforce CRM, Anaplan, Uncountable PLM, and Freshworks stored in Fabric
Process unstructured data from SharePoint, file systems, and document repositories outside Fabric
Implement RAG systems using FAISS, Pinecone, Weaviate, ChromaDB for hybrid structured/unstructured search
Build natural language interfaces for querying both lakehouse tables and external document sources
Create unified data processing pipelines combining Fabric data with external unstructured content
System Integration & Data Processing
Connect to Microsoft Fabric Data Lake using Delta Lake format and SQL endpoints
Access structured business data from all enterprise systems centralized in Fabric lakehouse
Integrate unstructured data sources: SharePoint documents, file servers, email archives
Process PDFs, Word docs, Excel files, images, and multimedia content from external systems
Implement real-time data streaming from Fabric Event Streams and external file monitoring
Build hybrid search capabilities combining Fabric structured data with external document vectors
Multi-Platform AI Development
Utilize OpenAI GPT-4, Anthropic Claude, Google Gemini, Meta LLaMA, and Cohere APIs Implement model routing and fallback strategies across AI providers
Build agents using LangChain, LlamaIndex, AutoGen, CrewAI frameworks
Deploy containerized solutions with Docker/Kubernetes for scalability
Required Skills:
Core Technical Expertise:
5+ years AI/ML development with enterprise data integration
3+ years Microsoft Fabric, Azure Data Lake, or similar lakehouse platforms
Advanced Python with AI/ML libraries (LangChain, LlamaIndex, Transformers)
SQL/KQL proficiency for complex data querying and analysis
Vector database expertise (FAISS, Pinecone, Weaviate, ChromaDB)
RAG system architecture and implementation at enterprise scale
AI & Machine Learning