AI Agents for Business Process Automation

Course overview

AI Agents for Business Process Automation is a practical professional training programme that equips organisations with the knowledge and skills to identify, design and implement AI-powered automation for business processes. Participants learn how AI agents can interpret documents and requests, retrieve organisational knowledge, make context-aware decisions, interact with enterprise applications, trigger workflows, communicate with users and complete multi-step tasks

The programme combines Agentic AI, Generative AI, workflow automation, RAG, APIs, enterprise applications and human-in-the-loop controls to demonstrate how organisations can redesign and automate real-world business processes.

Course objectives

• Understand AI-agent-based business process automation. Differentiate traditional automation, RPA, workflow automation and Agentic AI. Identify business processes suitable for AI-agent automation. Map and analyse existing organisational processes. Identify automation opportunities and bottlenecks. Design AI-agent workflows. Build AI agents capable of performing multi-step tasks. Connect agents to organisational documents and knowledge bases. Integrate AI agents with business applications and APIs. Automate document-intensive processes. Implement approval and escalation workflows. Apply human-in-the-loop controls. Monitor automated processes and agent actions. Measure automation performance and business value. Identify security and operational risks. Develop an organisational AI automation roadmap.

Who should attend

Course content

• Module 1: Introduction to AI-Powered Business Automation Business process automation Traditional workflow automation RPA Intelligent automation Generative AI AI copilots AI agents Agentic automation Autonomous vs semi-autonomous processes Where AI agents
• Module 2: Understanding AI Agents AI-agent architecture LLMs Instructions Reasoning and planning Memory Tools Function calling APIs Knowledge retrieval Actions Agent orchestration Practical: Build a simple AI agent capable of receiving a request and exec
• Module 3: Business Process Discovery Participants learn how to find processes that are suitable for automation. Topics: Process mapping Process owners Inputs and outputs Manual tasks Decision points Repetitive activities Bottlenecks Exceptions Approval
• Module 4: Selecting AI-Agent Automation Opportunities A practical framework for evaluating processes based on: Frequency Volume Complexity Repetitiveness Data availability Decision requirements Business value Risk Integration requirements Human involvem
• Module 5: Designing AI-Agent Workflows Participants learn to design: Trigger → Agent → Knowledge → Reasoning → Tool → Action → Validation → Approval → Completion Topics: Workflow triggers Agent tasks Decision points Conditional logic T
• Module 7: Enterprise Knowledge and RAG Organisational knowledge bases Document repositories RAG Embeddings Vector databases Retrieval Source grounding Permission-aware retrieval Knowledge-agent workflows Practical project: Create an organisational knowle
• Module 8: AI Agents + Workflow Automation Platforms Practical integration with technologies such as: Microsoft Copilot Studio Power Automate Microsoft 365 SharePoint Teams Dataverse Power BI Business applications Participants can create workflows such
• Module 9: AI Agents and APIs What is an API? REST APIs JSON Authentication API requests API responses Connecting agents to external systems Tool/function calling Webhooks Enterprise-system integration
• Module 10: Departmental AI-Agent Automation Participants explore practical use cases across departments. Finance Invoice processing Expense verification Financial report preparation Purchase requests Payment workflows Human Resources Recruitment workflo
• Module 11: Human-in-the-Loop Automation Not every decision should be fully autonomous. Participants learn how to implement: Approval gates Human review Escalation Confidence thresholds Exception handling Manual overrides Four-eyes approval High-risk ac
• Module 12: Multi-Agent Business Automation Introduction to multiple specialised agents.Topics: Agent roles Agent delegation Agent communication Supervisor agents Specialist agents Multi-agent orchestration Failure handling
• Module 13: Monitoring and Measuring Automation Participants learn to measure: Processing time Automation rate Error rate Human intervention Task completion Agent accuracy Cost per process Productivity SLA performance Business value
• Module 14: Security, Risk and Governance Data protection Access control Agent permissions Sensitive information Prompt injection Tool security API security Human oversight Audit trails Agent monitoring Responsible AI This module can naturally lead to you

Schedule & fees