Agentic AI for Project Managers

In this module, we explore where AI agents can support project work.

Use cases may include:

• Project status reporting
• Meeting summary and action tracking
• Stakeholder communication
• Market research
• Risk identification and escalation
• Requirement clarification
• Backlog refinement
• Sprint or iteration updates
• Decision support
• Knowledge search across project documents

Participants will learn how to separate useful agent use cases from risky, unclear, or low-value ideas.

Before building an agent, we need to design it.

This module introduces a simple design structure:

• Trigger: What starts the workflow?
• Input: What information does the agent receive?
• Context: What background knowledge does it need?
• Tools: Which applications or data sources can it use?
• Logic: What decision does it need to make?
• Output: What should it produce?
• Review: Where should a human approve or correct it?
• Log: What should be recorded for governance?

This gives project managers a clear way to design AI-enabled workflows even when they are not coding.

This is the core hands-on part of the program.

Participants will build a practical AI agent workflow using Make.com.

Example Build: Project Update Intelligence Agent

The workflow may:

• Receive a project update or input
• Classify the information
• Extract key risks, blockers, actions, and decisions
• Prepare a short summary
• Draft a stakeholder update
• Route the output for human review
• Save or send the final output based on approval

The purpose is to help participants experience how an AI-enabled workflow is created step by step.

This module demonstrates how AI agents can support common Google Workspace scenarios.

For example, a stakeholder may send an email asking for project information. The AI workflow can check the available context, prepare a draft reply, highlight uncertainty, and keep the project manager in control before anything is sent.

This helps participants understand how AI can reduce repetitive communication work without removing human judgment.

This module demonstrates how AI agent thinking can be applied in Jira and Confluence environments.

Examples may include:

• Reading Jira issue updates
• Summarizing project progress
• Identifying blockers
• Creating draft backlog items
• Checking Confluence knowledge before responding
• Preparing a sprint or project summary

The goal is to help participants see how AI agents can support delivery teams, agile teams, and PMO workflows.

AI agents need governance.

In project environments, we cannot allow agents to act without boundaries. This module focuses on practical controls.

Topics covered:

• Human-in-the-loop review
• Approval before communication
• Data privacy and access control
• Hallucination risk
• Logging and traceability
• Escalation rules
• When not to automate
• How to pilot an AI agent safely

Participants will learn how to think like responsible AI-enabled project leaders.

Participants will design an AI agent use case for their own project or PMO environment.

They will define:

• Problem statement
• Agent objective
• Inputs and data sources
• Tools required
• Workflow steps
• Human approval points
• Risks and guardrails
• Success measures

This exercise helps convert learning into a practical implementation idea.