Fellow Canadians: All inFlow plans are available in Canadian dollars 
Everyone’s talking about AI in operations, but almost nobody in manufacturing and supply chain actually knows where to start. In fact, in a survey of 400 operators, over 80% said they want to bring AI into their business, but they just don’t know whether it’ll actually pay off.
In the latest episode of Secret Life of Inventory, we sat down with Rehaan Bicha, co-founder and CEO of Metalinked, a manufacturing intelligence company built around automating the painful workflows that quietly eat your day. From quoting and sourcing to inventory and risk detection, Rehaan walked us through exactly how operators should (and shouldn’t) approach AI.
Watch the full episode below, or keep reading for the highlights.
Why most AI rollouts fail before they even start
Before you touch a single algorithm, Rehaan says there’s one question that matters more than any other: is your data actually ready? “A lot of times companies will come to us hoping AI will figure everything out,” he explains. But if your inventory data is scattered across an ERP, a stack of spreadsheets, and someone’s memory, no algorithm is going to save you. Step one is getting everything into a single pipeline before you even think about automation.
Two kinds of AI workflows (and how to tell them apart)
According to Rehaan, every AI use case falls into one of two buckets: it either finds an opportunity, or it makes an annoying task easy.
Opportunity-finding workflows look like identifying alternate suppliers, auditing your inventory, or spotting ways to cut down your SKU count. Annoyance-killing workflows are things like quoting and documentation, the tasks that quietly steal hours from your week.
His advice? Grab a piece of paper, brain dump every pain point on your plate, then rank them by ROI, whether that means cutting costs or saving time.
The $4 million inventory problem
Here’s where it gets interesting. Rehaan shared a story about a company in Los Angeles with roughly $4 million sitting in stock-keeping units and no clear picture of what was connected to what. Once his team organized the data, the AI started flagging components that could be reused across production lines instead of being reordered from scratch.
The result? Around $600,000 in savings, nearly triple what the company originally expected.
The vibe-coding trap manufacturers need to watch for
With AI tools making it easier than ever to build your own software, more companies are “vibe coding” internal tools instead of buying a solution. Rehaan admits this democratizes AI, but it’s a double-edged sword. Without the right infrastructure and security in place, these tools can rack up hidden on-demand costs or leave sensitive supplier and customer data exposed.
Three mistakes companies make when they start
When asked what trips companies up most, Rehaan didn’t hesitate:
- Assuming AI will run the whole business on autopilot
- Skipping the “context” layer, meaning the tribal knowledge that makes your business your business
- Treating data security as an afterthought
Miss any one of these, and you’ll end up with an AI tool that’s generic at best, and a liability at worst.
Questions to ask before you buy
If you’re evaluating an AI vendor, Rehaan says to lead with the boring logistical questions before getting excited about ROI. Where is your data going? Is the pipeline secure? Is the vendor SOC 2 compliant? Can the solution run on your own premises?
Data protection comes first. Many AI solutions run externally on the cloud, so it’s worth confirming the pipeline is secure and that the vendor holds SOC 2 compliance, the standard most AI companies need for handling other businesses’ sensitive data. If you’re in a regulated space like government, defense, or aerospace, on-premises deployment often isn’t optional, it’s a requirement.
Only once those boxes are checked should you move on to the fun part: the actual ROI.
The bottom line
AI isn’t going to replace the handshake deals and relationships that make manufacturing and supply chain work. But it can absolutely take the redundant, annoying tasks off your plate and surface opportunities you didn’t know existed.
Want the full conversation, including why smaller manufacturers have a speed advantage over larger ones? Watch the complete episode of Secret Life of Inventory.
0 Comments