I am trying to develop some Proof Of Concept on use of Agentic AI for Project Management. Most of the AI in Project Management discussions are around use of Generative AI in Project Management, but I'd like to take it a step further and build some workflows.
Any suggestions?
Saving Changes...
Sort By:
Faisal Ahmed RonyFounder & Chief Editor| Total InfoHubDhaka, Bangladesh
Love this approach. Generative AI is great for drafting summaries, but Agentic AI is a game-changer for active orchestration.
A highly valuable PoC would be an Autonomous Stakeholder Reporting & Escalation Agent. It can scan daily updates, determine the severity of risks using predefined project constraints, and independently manage the communication workflow—sending different levels of summaries to the team vs. executives based on priority. Best of luck with your PoC development! Saving Changes...
Luis BrancoCEO| Business Insight, Consultores de Gestão, LdªCarcavelos, Lisboa, Portugal
A good starting point might be to choose workflows where the agent can observe, act within clearly delegated boundaries, and escalate when predefined conditions are reached.
A few PoC candidates:
Risk sensing agent: monitors meeting notes, issues, changes and status updates, then identifies emerging risks, updates triggers and initiates predefined review workflows.
Decision follow-up agent: tracks decisions, owners, due dates and dependencies, then follows up and escalates when accountability or execution becomes unclear.
Change impact triage agent: gathers relevant information for a change request, identifies missing inputs and prepares an initial impact view across scope, schedule, cost, risks, benefits and stakeholders.
Stakeholder engagement agent: detects early signals of misalignment or low engagement from project communications, within explicit privacy and ethics boundaries.
Lessons learned agent: identifies recurring patterns during the project and surfaces relevant learning when similar situations reappear, not only at closure.
The key design question may be less “what can the agent do?” and more “what may be delegated, within which authority boundaries, under what escalation conditions, and with what accountability?” Saving Changes...
