AI Support Agent with Azure and Copilot Studio
Budget / Salary₹600–1,500
TypeFreelance project
LocationRemote
Posted1 hour ago
• Designed and configured AI agents using Microsoft Copilot Studio to assist support teams with incident resolution and operational troubleshooting.i need aproject training for the below• Built an end-to-end RAG architecture involving document ingestion, embedding generation, vector search, retrieval, and LLM response generation.
• Implemented Azure OpenAI embeddings for converting knowledge articles, runbooks, problem records, and support documentation into vector representations.
• Utilised Azure AI Search and vector database capabilities to perform semantic search using embeddings, metadata filtering, and Top-K retrieval techniques.
• Created metadata-driven search filters using application, environment, severity, and support-group attributes to improve answer accuracy.
• Developed conversational AI prompts and prompt templates for support workflows, troubleshooting guidance, and knowledge retrieval.
• Built Python-based ETL pipelines to ingest documents from SharePoint, knowledge repositories, CSV files, and operational databases.
• Developed automation workflows using n8n to orchestrate document ingestion, embedding generation, notification workflows, and incident enrichment processes.
• Created multi-step AI workflows integrating retrieval, reasoning, recommendation generation, and response validation.
• Implemented AI governance controls including source attribution, confidence checks, prompt guardrails, and response validation mechanisms.
• Configured semantic search and vector similarity retrieval utilising cosine similarity and Top-K search techniques.
• Designed AI chatbot experiences capable of answering operational support, incident management, and infrastructure-related questions.
• Collaborated with business stakeholders to gather requirements and translate support challenges into AI-driven solutions.
• Developed reusable solution documentation, architecture diagrams, technical runbooks, and operational support guides.
• Conducted user demonstrations, knowledge-sharing sessions, and AI adoption workshops.
• Monitored retrieval accuracy, prompt effectiveness, and operational costs to optimise AI agent performance.
• Implemented secure handling of sensitive information through metadata filtering and role-based access considerations.
• Implemented Azure OpenAI embeddings for converting knowledge articles, runbooks, problem records, and support documentation into vector representations.
• Utilised Azure AI Search and vector database capabilities to perform semantic search using embeddings, metadata filtering, and Top-K retrieval techniques.
• Created metadata-driven search filters using application, environment, severity, and support-group attributes to improve answer accuracy.
• Developed conversational AI prompts and prompt templates for support workflows, troubleshooting guidance, and knowledge retrieval.
• Built Python-based ETL pipelines to ingest documents from SharePoint, knowledge repositories, CSV files, and operational databases.
• Developed automation workflows using n8n to orchestrate document ingestion, embedding generation, notification workflows, and incident enrichment processes.
• Created multi-step AI workflows integrating retrieval, reasoning, recommendation generation, and response validation.
• Implemented AI governance controls including source attribution, confidence checks, prompt guardrails, and response validation mechanisms.
• Configured semantic search and vector similarity retrieval utilising cosine similarity and Top-K search techniques.
• Designed AI chatbot experiences capable of answering operational support, incident management, and infrastructure-related questions.
• Collaborated with business stakeholders to gather requirements and translate support challenges into AI-driven solutions.
• Developed reusable solution documentation, architecture diagrams, technical runbooks, and operational support guides.
• Conducted user demonstrations, knowledge-sharing sessions, and AI adoption workshops.
• Monitored retrieval accuracy, prompt effectiveness, and operational costs to optimise AI agent performance.
• Implemented secure handling of sensitive information through metadata filtering and role-based access considerations.
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.