Food distribution is not known for large IT teams, with most mid-sized distributors operating with lean support staff. While capital tends to go elsewhere (trucks, warehouse equipment, drivers, and inventory), technology teams often consist of just a handful of people responsible for everything from networks to ERP systems to integrations.

At Brown Foodservice, that reality is especially clear. The company runs a two-person IT operation supporting 180 employees. Yet despite the limited resources, the team is building integrations, automating workflows, and modernizing legacy systems using AI.

Their approach illustrates how AI can function as a practical operational tool inside a distributor, rather than just another piece of technology hype

A 29-Year IT Veteran Meets AI

Chris Fissler has been working at Brown Foodservice for 29 years. He’s a self-taught programmer responsible for nearly everything on the data and systems side of the business. 

He works alongside 

Like many distributors, Brown’s core ERP system is decades old. As Chris put it, “Our ERP system is the best the nineties has to offer.”

Legacy systems present a familiar challenge in food distribution: they’re stable but difficult to integrate with modern tools. Rather than replace the ERP, Chris began experimenting with ChatGPT and Microsoft Copilot to build bridges around it.

Initially, AI served as a coding assistant. He would paste functions into ChatGPT or Copilot and ask questions like:

• Why isn’t this code working?
• Can this function be rewritten more efficiently?
• Where is this bug coming from?

The results were immediate.

“It’s improved my confidence with programming, and it’s made me a lot faster with the things I was already familiar with. There are definitely more things I’m not afraid to try now.”

What previously required hours of debugging or research could often be solved in minutes.

Building a “Data Bus” for a Legacy ERP

The most ambitious project Chris tackled was creating what he calls a “data bus”, an integration layer designed to connect the company’s ERP with modern applications.

Before starting, he asked ChatGPT to help architect the system. “I told it what I wanted to build and asked it to develop a blueprint - what it came out with helped me see how fast it could get done”

The result was a service layer built as a Node.js Express server that sits between the ERP and external systems, allowing Brown Foodservice to integrate modern tools without modifying the ERP itself.

The system now handles several responsibilities:

• Managing vendor integrations
• Receiving and processing webhooks from external platforms
• Tracking incoming systems and data sources
• Syncing data across applications
• Routing updates back into the ERP

In effect, the data bus acts as a modern integration layer around a legacy ERP, allowing new systems to connect without changing the core platform.

Eliminating Manual Routing Workflows

The first major use case for the data bus was route optimization.

Brown Foodservice uses Route4Me for delivery routing, but previously the system required manual coordination between the routing platform and the ERP.

The integration Chris built now synchronizes them: If routing changes occur in Route4Me, the ERP updates automatically, and If data changes in the ERP, Route4Me reflects those updates.

Orders can now:

• Route automatically
• Update in near real time
• Flow directly into warehouse palletizing

The project removes the need for intermediary processes and manual reconciliation between systems. Like Chris says, “It’s about as close to real time as I can get it.”

Compressing Months of Research Into Weeks

For Brown’s COO, Daniel Neeley, the biggest impact isn’t just automation. It’s how much faster the team can build things.

Historically, researching and implementing a new technology might take three to six months, Now it can happen in weeks.

“Being able to get educated on something completely new and build a working bridge in a matter of weeks - that just wasn’t possible before. AI dramatically shortens the learning curve for unfamiliar technologies.”

Instead of reading documentation for weeks, developers can iterate quickly through architectural ideas, sample code, troubleshooting, and even get into implementation detail. This compresses experimentation cycles and enables small teams to attempt projects that previously required larger IT departments.

Digitizing Safety Processes in Two Weeks

AI isn’t only helping with integrations, the team recently used it to build a warehouse equipment inspection system.

Food distributors operating forklifts and material handling equipment must track safety checks for OSHA compliance. Many companies still handle this with paper logs, which was what Brown was doing until the beginning of 2026 when within two weeks, the IT team created a digital replacement.

Warehouse operators now complete a pre-trip inspection on a kiosk, recording:

• Operator identity
• Equipment condition
• Inspection approval

The system writes the data into a database automatically.

Previously, inspections were tracked on paper and accountability was difficult to scale with any reporting requiring manual effort. Now the process is digital and searchable. As Daniel describes it:

“You can’t have a safety culture without accountability. And without digital tools, there’s no scalability to accountability.”

Preparing for AI With a Data Lake

Beyond immediate automation projects, Brown is also preparing for more advanced AI use cases. The company has begun aggregating operational data into a centralized data lake containing all order and product data alongside operational and historical business data.

The long-term goal is to build a natural language interface to company data.

Instead of asking IT to generate reports, leaders could ask questions directly.

For example:

• “Show me sales by route last week.”
• “What products are trending with independent restaurants?”
• “Which trucks are underutilized?”

The system would generate reports automatically. “Someone like Daniel could ask for whatever report he wants” Chris reported, “and it would create it without me programming it.”

A Small IT Team Supporting 180 Employees

Despite these initiatives, Brown’s technology resources remain extremely lean.

The company’s IT team consists of:

• Chris Fissler (data systems and integrations)
• Tyler Blackburn (network and infrastructure specialist)

Together they support 180 employees across operations, warehouse, and sales.

This staffing model is typical in food distribution, where support departments are often minimal compared to operational staff. “You typically run with a skeleton crew in this industry.”

AI changes the equation.

Instead of hiring a large development team, the company can prototype ideas quickly, automate small internal tools, and build integrations incrementally. AI becomes a force multiplier for small technical teams.

The Next Skill: Prompting

The conversation at Brown reflects a broader shift in how technical work is done. Twenty years ago, a critical skill in IT was knowing how to search the internet effectively. Like Chris says, “I used to say that I’m not in IT because I’m better at computers. I’m in IT because I can Google better than you.”

Now that skill is evolving into something new: Prompting AI tools effectively is becoming its own discipline.

“Googling was a skill you had to learn. Now the skill is knowing how to work with AI prompts.”

This ability is increasingly important not just for developers, but for anyone working with technology.

‍