Flexport has started an internal 90-day programme in which employees from different departments learn to build their own AI tools.
The approach behind it matters: teams learn to describe their own work problems well enough to build small automations that can be tested, without waiting for a large system to be completed elsewhere.
Many logistics AI use cases arise where people encounter the same friction every day:
- A form is almost correct.
- Customs or transport information is missing.
- An email contains a crucial detail.
- A permit needs reading.
- A follow-up question arrives too late.
- A customer asks for a status that exists somewhere, but is unavailable where it is needed.
The people familiar with these problems need to learn how to formulate useful tasks for AI. They do not all need to become programmers.
We see helping teams describe their work clearly as a new management responsibility.
AI know-how for logistics professionals in 2 minutes
What is a skill?
A skill is a reusable capability for an AI system.
A prompt often covers a single task: “Read this document.” A skill defines a repeatable approach: “When a transport permit arrives, check these points, extract these fields, flag uncertainties and prepare these follow-up questions.”
Skills make AI work more repeatable. In logistics, they might cover:
- Reading a permit.
- Extracting conditions.
- Checking transport data.
- Flagging missing information.
- Preparing a question for dispatch.
A set of small, carefully managed capabilities makes AI practical.