Published May 26, 2026 in Announcements
Reimagining Supply Chain Resilience

Overview
In the world of supply chain management, we have spent the last ten years focusing on Visibility. We built digital dashboards to see problems as they happen. But here is the hard truth: Visibility without action is just a front-row seat to a disaster.
I am currently conducting research to bridge the gap between knowing a problem exists and solving it automatically. I am doing this not only for my startup but also to provide academic backing that contributes to the knowledge body for both entrepreneurs and researchers. My goal is to move beyond simple monitoring toward Agentic AI.
What makes Agentic AI different?
New section title
Traditional software is reactive, it waits for a person to click "approve" or "re-route." Agentic AI shifts this toward autonomous resilience.
☑️ Goal-Oriented Reasoning: Instead of just following "if-then" rules, AI agents are given a target. For example: "Maintain tea export schedules to the Colombo port despite local transport strikes or fuel shortages." The AI then finds the best alternative path to hit that goal.
☑️ Autonomous Negotiation: Imagine AI agents representing different garment factories in Sri Lanka "talking" to each other to share fabric inventory or shipping containers without thousands of manual emails.
☑️ Adaptive Learning: These systems don't just solve a problem once. They learn from past port delays or currency fluctuations to make the supply chain "harder" against the next crisis.
The Focus:
I am studying the adoption frameworks of these technologies. I want to understand how we move from "Human-in-the-loop" (doing the work) to "Human-on-the-loop" (supervising the AI). This is vital to ensure that autonomous decisions align with Environmental, Social, and Governance (ESG) goals.
Building a resilient supply chain is not just about having more data. It is about having the agency to act on that data in milliseconds, not days.
I’m curious to hear from my network, What is the biggest hurdle you see in letting AI agents take "autonomous actions" in your procurement or logistics workflows?


