icon icon

Copilot Studio – designing and deploying your own AI agents

icon

Training process

Training needs analysis

If you have specific requirements regarding the training programme, we will carry out a training needs analysis for you. This will guide us on which aspects of the programme should receive greater emphasis, so that the training programme meets your specific needs.

What will you gain?

icon

Confident tool selection - You will clearly distinguish Copilot Pro, Microsoft 365 Copilot, and Copilot Studio, so you can choose the right option for your team, business process, and rollout scope.

icon

Agent with a clear purpose - You will learn how to define the agent’s role, response scope, and boundaries, so you can build solutions that genuinely support users in everyday tasks and requests.

icon

Stronger conversation flows - You will design topics, questions, decisions, and branches on your own, helping you create conversations that stay consistent, logical, and resilient to common user mistakes.

icon

Use of company knowledge - You will learn how to connect SharePoint, OneDrive, and PDF or Word documents, so your agent can use up-to-date materials and answer based on actual organizational content.

icon

Process integrations - You will explore how Copilot Studio works with Power Automate and external APIs, allowing you to extend the agent with data retrieval, workflow triggers, and task automation.

icon

More reliable testing - You will practice testing the agent in draft mode, handling exceptions, and fixing broken paths, so you can publish solutions that behave more reliably and predictably.

icon

Ready for publishing - You will learn how to publish your agent in Microsoft Teams, on websites, and in other apps, while also configuring access so the solution reaches the right users safely.

icon

Improvement through analytics - You will understand how to review conversation history and response effectiveness, so you can improve the agent iteratively and remove issues that weaken the user experience.

Training programme

1. Introduction to Copilot Studio

  • what Copilot Studio is and how it works,
  • differences between Copilot Pro, Microsoft 365 Copilot and Copilot Studio,
  • examples of AI agent applications in companies,
  • overview of the user interface and tool structure.

2. Planning and designing an AI agent

  • defining the agent's goal and functions,
  • designing the conversation logic,
  • creating topics and trigger phrases.

3. Building conversation scenarios

  • types of conversational blocks: questions, answers, decisions, actions,
  • conditions, branching and conversation flow logic,
  • use of variables and context management,
  • handling exceptions and erroneous paths.

4. Integration with data and documents

  • connecting knowledge sources (SharePoint, OneDrive, PDF files, Word),
  • using the RAG (Retrieval-Augmented Generation) approach,
  • integration with Power Automate and external services (API).

5. Testing and publishing the agent

  • testing the agent in working mode,
  • publishing in Microsoft Teams, on a website, in external applications,
  • configuration of availability and user management.

6. Analysis of operation and development of the agent

  • review of conversation history and analysis of effectiveness,
  • searching for and eliminating errors,
  • iterative development and updating of the agent.

7. Practical task – creating your own agent

  • designing a conversation scenario according to the selected case,
  • building the agent, tests, sharing,
  • presentation and discussion of the results.

What are the prerequisites for participating in the training?

icon

Microsoft 365 basics - You should be comfortable using Microsoft 365 apps and services, including Teams, OneDrive, and SharePoint, so the integration examples covered in class are easy to follow.

icon

Process thinking - You should be able to describe a simple business process step by step, because during the training you will design conversation logic, decisions, conditions, and actions.

icon

Working with documents - You should have basic experience working with documents and information sources, so you can efficiently prepare the content your agent will use when generating answers.

icon

No-code basics - A general understanding of no-code or low-code tools will help you, because during the training you will configure logic, integrations, and publishing without coding.