Marlow & Finch Logistics manages freight across the UK, EU and Middle East, and their operations team was manually triaging shipment exceptions from six disconnected tracking systems.
We built a multi-agent system where a Monitoring Agent watches live shipment data for anomalies, a Prediction Agent forecasts likely delays using historical patterns, and a Communications Agent drafts proactive customer updates — with a human operations lead approving anything above a defined risk threshold before it goes out.
The agents coordinate through a shared task queue built on LangGraph, with full traceability into why each decision was made, which mattered enormously for gaining the operations team’s trust in the system’s first months.
Tell us about your workflow and we'll tell you honestly whether an AI agent is the right next step.
Start the Conversation