Small and mid-size manufacturers face new hurdles as digital transformation and AI advance, but practical, business-focused solutions are within reach. In this episode, Ira Sharp speaks with Patrick Byrne, cofounder of Annora AI, an engineer-turned-data scientist specializing in manufacturing efficiency. Together, they address the need to find real bottlenecks, use both human and machine data to pinpoint solutions, and provide actionable tips on getting started with AI—no matter your company’s size. Their discussion delivers clear guidance for manufacturers seeking lasting impact, rather than quick fixes.
Pinpointing Bottlenecks: The Importance of Finding the Right Problems
Patrick Byrne emphasizes that technology, no matter how advanced, cannot fix what hasn’t been fully understood. He shares that Annora AI’s approach starts by deeply listening to teams across an organization, identifying the most pressing pain points that ripple across departments. Byrne notes, “No one understands the problems in a department better than the department leader. They hear it from all of the individual people that report to them, and they can feel that pain, and they see how it affects other parts of the company.” By mapping these pain points—often caused by disconnected systems and “digital waste” where important information is hidden or duplicated—Annora AI helps manufacturers clarify where the real issues begin.
Their focus is on ensuring improvements chase real, quantifiable problems, not symptoms. As Byrne explains, “If you’re trying to solve the wrong problem, it doesn’t matter how good your solution is.” This process not only saves time and cost, but also builds internal support for any solutions that follow.
Combining Employee Insights and Data: A Double Lens Approach
Rather than relying on digital systems alone, the Annora AI team prioritizes conversations with staff to unearth hidden workflow inefficiencies and then supplements these findings with hard data from ERPs and machine systems. Byrne describes how they conduct interviews, document everything, and use their AI tools to analyze recurring pain points across functions—from sales to engineering and project management. This approach ensures that both human stories and numerical data drive improvement efforts.
Byrne points out that even with the growth of automation, “You do need to look at this machine data or this ERP data or the company data itself to really be able to quantify these things. But we want to start with the understanding that the department manager feels this problem.” Capturing staff perspectives builds trust, aids adoption, and sharpens focus, while machine data measures the true size and impact of each identified problem.
Getting Started with AI in Manufacturing: A Roadmap for Progress
One major concern for manufacturers is knowing how to begin their digital journey without huge budgets or risking operations. Byrne offers a straightforward entry plan: start with internal conversations to agree on the core problems to solve. Most often, he says, these problems relate to information retrieval—finding critical documents or knowledge buried across many locations—as well as data visualization and automation of repetitive tasks.
“Manufacturers, especially like, a lot of these smaller manufacturers...have got all this information spread across all these different systems, and they’re not able to find it,” Byrne comments. Introducing simple information-retrieval systems or dashboards can save hours across teams. Byrne highlights that such entry projects are no longer expensive or high-risk, thanks to AI tools that rapidly build practical solutions. He advises, “You don’t need to make any investments, and you don’t need to really put any big money into this. I think you should just be considering your options and looking at where things are going because AI is moving so fast that if you don’t get on now, it’s going to be a lot harder to catch up in, you know, six months or a year.”
Key Quote From The Episode
“If you’re trying to solve the wrong problem, it doesn’t matter how good your solution is. So we need to find what the problem is, and then we can look at, you know, how do we solve that using AI or other engineering type systems or data systems.” – Patrick Byrne
Key Takeaways
- [00:04:47] Byrne stresses the need for honest internal discussions to surface real bottlenecks, involving voices from across the company.
- [00:07:20] Anora AI’s process integrates staff interviews with machine data to locate and quantify workflow issues, ensuring results are both relevant and measurable.
- [00:13:35] Byrne encourages companies to begin with small, low-risk projects focused on improving access to information, visibility, and workflow automation, setting a foundation for future high-impact AI applications.
Wrap Up
Manufacturers seeking to improve operations and stay competitive should begin by identifying their most costly process bottlenecks through direct conversations and careful analysis. By blending employee expertise with digital data, companies can develop targeted, practical solutions. Byrne’s advice: start with simple, high-impact projects like smarter information searches or dashboards, and build steadily from there. To take action, leaders should schedule internal meetings focused on pinpointing core problems, invite cross-department input, and consider engaging partners who can quickly deliver tailored solutions using modern AI tools.
About the Guest
Patrick Byrne is cofounder at Annora AI, drawing on his background as a mechanical and manufacturing engineer and work at Intel to help manufacturers improve efficiency, lower costs, and grow profits. His hands-on, business-oriented approach combines technical expertise with a focus on real-world results for small and mid-size manufacturers.
