Highlights
- Build and deploy agents using Copilot Studio
- Design effective multi-turn conversations with topics, triggers, and variables
- Use generative AI with grounded enterprise data
- Integrate with Power Automate and external data sources (SharePoint, APIs, SQL)
- Apply orchestration for multi-step and goal-driven tasks
- Implement logic, conditions, and error handling
- Evaluate, test, and improve agent performance
- Apply governance, guardrails, and responsible AI practices
Course Details
Core Copilot Studio Foundations
Build and deploy agents using Copilot Studio
- Copilot Studio interface walkthrough
- Creating your first agent
- Understanding structure and workflow
Design effective multi-turn conversations with topics, triggers, and variables
- Core concepts: topics, triggers, entities, variables
- Multi-turn conversation design principles
Use generative AI with grounded enterprise data
- Generative answers (modern capability)
- Grounding with enterprise data (SharePoint, files, websites, Dataverse, etc.)
Apply orchestration for multi-step and goal-driven tasks
- Introduction to generative orchestration (how copilots plan tasks)
Evaluate, test, and improve agent performance
- Basic testing & debugging
Apply governance, guardrails, and responsible AI practices
- Data governance & responsible AI considerations
Hands-On Labs
- Build a simple “Ask the Organisation” FAQ agent
- Add grounded generative answers using internal documents
- Create topics, add triggers, test and refine the conversation flow
Integrations, Logic, Orchestration & Evaluation
Integrate with Power Automate and external data sources (SharePoint, APIs, SQL)
- Power Automate integration (triggering flows, passing data)
- Using connectors (Dataverse, SharePoint, SQL, APIs, etc.)
- Reading/writing external data safely
Implement logic, conditions, and error handling
- Advanced conversation design (branching, conditions, variables)
- Entity extraction & data capture
- Error handling & fallback strategies
Apply orchestration for multi-step and goal-driven tasks
- Generative orchestration deep dive
- How copilots plan and choose actions
- When to use orchestration vs. topic flows
- Autonomous agent behaviours
- Goal-driven execution
- Multi-step reasoning
Evaluate, test, and improve agent performance
- Agent evaluation
- Creating evaluation datasets
- Running automated tests
- Improving accuracy and reliability
- Analytics & conversation transcripts
Apply governance, guardrails, and responsible AI practices
- Guardrails and predictable behaviour
Build and deploy agents using Copilot Studio
- Deployment options (Teams, web, channels) & adoption considerations
Hands-On Labs
- Build a scenario-based agent (service request, knowledge assistant, status lookup)
- Connect to a data source and return real values
- Add orchestration to complete a multi-step task
- Run an evaluation and improve the agent
- Test, debug, refine
Who should attend
- Business analysts and process owners looking to automate workflows with intelligent agents
- Developers and technical users integrating agents with APIs, data sources, and enterprise systems
- Power Platform users wanting to extend Copilot and Power Automate capabilities
- Solution architects designing scalable, AI-driven business solutions
- Teams responsible for deploying, testing, and improving conversational or agent-based systems
Feedback
4.8 out of 5 average
"Our tailored course provided a well rounded introduction and also covered some intermediate level topics that we needed to know. Mariella gave us some best practice ideas and tips to take away. Brian Leek, Data Analyst, May 2024
“JBI did a great job of customizing their syllabus to suit our business needs and also bringing our team up to speed on the current best practices. Mariella was excellent & contents were very impressive” Brian F, Team Lead, RBS, Data Analysis , 20 April 2024