I've been in logistics IT since 1999. I've worked inside freight forwarders. I've built TMS systems. And for years, I've had a conversation I absolutely hated having.
A customer would call. We'd missed something in their shipment. Maybe we shipped something they explicitly told us not to ship. Maybe documentation was incomplete. Maybe a compliance requirement was overlooked.
And I'd have to explain why.
"We'll add it to the daily checklist."
"We'll double-check the report every day."
"We'll make sure the team reviews it manually."
I hated saying those words because I knew we couldn't follow through. Not reliably. Not at scale.
The Tools Couldn't Keep Up With Reality
Here's the problem: customers have requirements that don't fit neatly into TMS rules.
"We only ship consumer electronics—things you'd find in a Best Buy."
How do you write that as a database query? You can't. At least not with the tools we had.
Or a customer would have a specific routing restriction, or a packaging requirement, or a documentation rule that applied only to certain products going to certain countries on certain carriers.
We could build rules for the simple stuff. Port codes. Required fields. Date validations. But the complex, abstract requirements? The ones that required judgment? Those lived in people's heads.
And people are busy. Files get bigger. Teams get stretched. Things slip through.
I'd sit in those customer calls knowing the real answer was: "We don't have a way to enforce this without assigning someone to manually review every shipment. And we don't have the resources to do that."
But you can't say that. So we'd promise to be more careful. And we'd miss things again.
Building Integrations at Chain.io
When I started Chain.io, we focused on connecting systems. TMS to customs software. Forwarders to carriers. Making data flow automatically between platforms.
And that work matters—those integrations move thousands of shipments every day.
But even with perfect data flows, teams were still stuck with the same problem I'd faced: finding the problems hiding in the spreadsheets.
Your customer sends you a 500-row file. Three shipments violate their requirements. Which three?
Your forwarder sends you an updated file every morning. 520 rows. What changed? What needs attention?
The answer was always the same: manual review. Scroll through the rows. Cross-reference with yesterday's file. Hope you catch everything.
We Built What I Wished I Had
AI Checks does what I needed back then but couldn't build: it reads your spreadsheet, applies rules written in plain English, and finds the problems automatically.
"Flag any shipment going to Brazil."
"All products should be things you'd find in a Best Buy store."
"We only accept FOB incoterms."
You write the rules the way you'd explain them to a person. AI Checks applies them to every row. And when you upload an updated spreadsheet tomorrow, it shows you only what changed.
A shipment you reviewed and archived on Tuesday? It comes back if new data reveals a problem. The system remembers. It tracks. It doesn't make you start over every day.
No formulas. No manual cross-referencing. No promises you can't keep.
This Solves the Customer Call Problem
The hardest part about those calls wasn't explaining what went wrong. It was knowing we didn't have a real solution.
Now I can say: "Write your requirements in plain English. We'll apply them automatically. If something violates your rules, you'll know before it ships."
That's the conversation I wanted to have 15 years ago.
This is Practical AI
I don't care about AI that forces you to change how you work. I care about AI that solves problems you have today with tools you already use.
You have spreadsheets. You have requirements that are hard to enforce. You have shipments that need review.
AI Checks takes those three things and makes them work together without requiring an IT project, a new system, or restructured data.
That's it. That's the product.
Try It Yourself
We built an interactive demo with sample data. You can see how it works without uploading anything.
When you're ready, upload your own spreadsheet. If it helps, keep using it. If it doesn't, you'll know in five minutes.
Brian Glick is the CEO and founder of Chain.io. We move the data that moves your freight.
