DarwinAI is making waves in just about every industry.
First, they were helping Audi develop far more efficient autonomous driving technology. Then they started helping Lockheed Martin with their artificial intelligence applications. When COVID-19 spread across the world, they applied their Explainable AI (XAI) platform toward developing unique new ways to detect positive cases.
And now, DarwinAI is becoming the leader in applying machine learning to challenges in the manufacturing sector, an industry that is ripe with opportunities to innovate with new technologies.
Who are they helping? What types of problems are they solving? How can companies looking to innovate work with them?
We asked Sheldon Fernandez, the CEO at DarwinAI, to answer these questions and more.
Q: Let’s start with something that’s pretty central to this whole discussion – what is explainable AI?
FERNANDEZ: Explainable AI (XAI) is essentially a collection of technologies that reveals how and why AI makes the decisions that it does. This capability is important for optimizing performance and building trustworthy applications. Darwin’s XAI technology applies to a wide range of automation and human-in-the-loop decision-making use cases, which enables us to build superior enterprise solutions that have a smaller memory footprint, [are] more computationally efficient and perform with high levels of precision.
Q: When manufacturers come to you, what kinds of problems are they usually trying to solve? Are they looking to improve processes, products or both?
FERNANDEZ: The problems that we encounter most often from manufacturing clients relate to visual quality inspection. Visual quality inspection can be a long and tedious process that manufacturers struggle with due to missed defects, the need for more lead time and high rework costs. It is one of the most crucial steps in the production process as quality inspection can prevent and limit defects before a product hits the shelf. Darwin’s Explainable AI helps manufacturers improve both processes and product quality by locating defects and implementing continuous AI learning to make the solution smarter and more accurate over time.
Q: Is there a particular type of manufacturing that DarwinAI’s platform is best suited to help? Automotive? High tech? Or is the platform flexible to potentially help any company?
FERNANDEZ: DarwinAI can help customize a solution for any manufacturer in need of visual quality inspection. We help clients from industries such as aerospace, automotive, consumer goods, electronics and healthcare. Some examples of manufacturing solutions we’ve devised include scratch detection, PCB inspection and melt pool inspection.
Our mission is to help manufacturers adopt AI they can trust. In order to achieve this, we deliver robust and best-in-class quality inspection solutions with a relentless focus on customer success so manufacturers can transform their business using this technology.
Q: What is the first step for a company that wants to investigate using artificial intelligence?
FERNANDEZ: Many manufacturers have tried to implement AI on their own. They typically have challenges with resources and figuring out how to integrate an AI system into their production process. Many consult with us as a first step. We help them get started by assessing their production process, taking the time to understand their business objectives and what they want to achieve by adding AI to their production processes. We take the time to recommend and build the best AI solution to get results they can see by improving their production KPI’s.
Q: What does working with DarwinAI look like? Is it an ongoing partnership or do you simply provide a solution and then disappear until they need you again?
FERNANDEZ: Because of our unique technology, we build XAI models that require less data, and are more accurate in helping manufacturers realize quicker time to value. We ensure the XAI solution is designed and implemented correctly and guide success throughout the entire process leading to improved product quality and manufacturing KPIs. We take a very collaborative approach to working with our manufacturing clients. Our Customer Experience team is very responsive to any needs or issues that may come up during the entire implementation process.
Q: While manufacturing is embracing technology more than ever – particularly in the form of automated processes – there’s still a ton of manufacturing leaders who are hesitant to incorporate new technologies. How would you convince them to explore something like AI?
FERNANDEZ: Manufacturers are constantly striving to improve the efficiency in their production process. Our visual quality inspection platform can be trained with significantly less data than our competitors, before we deliver a working system. Contrary to the misconception that AI eliminates jobs, our inspection process still requires quality inspectors to validate predictions. DarwinAI’s solution is about making the operator’s job more efficient too.
Q: Last, we’ve hardly talked about Waterloo, but DarwinAI was born here and one of your senior leaders still works at the University of Waterloo. Can you tell us a little bit about what makes Waterloo such an exciting place for companies working in artificial intelligence?
FERNANDEZ: The University of Waterloo is one of the most innovative universities in Canada with an excellent co-op program, as well as state-of-the-art facilities such as Velocity and the Waterloo Artificial Intelligence Institute. It’s no surprise that Waterloo is a tech hub for AI companies since there are so many intelligent people and opportunities to expand here. Seeing students being just as passionate as us also brings a new and fresh perspective to the table when we are solving challenging problems using XAI.