Last year, Mechanical Engineering professor Andrew Gadsden collaborated with NASA to send a robotic, autonomous telescopic mount above 95 per cent of the Earth’s atmosphere.
So it’s only fitting that he’s now working on a new partnership with a Toronto-based data science and artificial intelligence (AI) consulting firm called Adastra — Latin for “to the stars.”
Funded through a four-year, co-sponsored grant supported by NSERC Alliance and Mitacs Accelerate, Gadsden’s project is exploring two rapidly evolving areas of AI — explainable AI and agentic AI — with the goal of creating intelligent systems that are not only capable, but also transparent and reliable.
A black box no more
One phase of the project involves working with agentic AI to deploy reliable systems in a variety of industries, including health care, finance, manufacturing and even outer space.
Agentic AI is an autonomous system that operates with limited human supervision and is responsible for making decisions and adapting to changing environments.
An example of this type of system, explains Gadsden, would be automated fraud detection. An AI agent could be deployed to flag suspicious charges as fraud, collect and gather relevant supporting information about fraud, compare similar claims, and automate audit report generation.
With limited knowledge of how the system works and minimal human interactions with it, however, how can users trust that the agent has correctly identified the charge as fraud?
This is where explainable AI (XAI), the second phase of the project, comes in.
XAI seeks to build credibility and trust in AI systems among users. Current AI models incorporate something called a black box, where users can see what goes into the system (inputs) and what comes out of it (outputs), but not the inner workings that make the system function.
By being able to see into the black box, explains Gadsden, the user gains a better understanding of the overall system, has more confidence in what the AI model is telling them and can regularly monitor the health of the system. Furthermore, increased transparency supports building more robust and accurate solutions as XAI shows the functional boundaries of the underlying models, tools, and agentic processes.
“The future of AI isn’t just about smarter machines,” says Gadsden. “It’s about building systems we can understand, rely on and responsibly integrate into the world around us.”
A growing ecosystem here at home
For Gadsden, this work with Adastra is more than just a single research project — it reflects a larger picture that’s forming across campus.
At McMaster, he explains, researchers are helping to shape how AI is developed and deployed beyond the university, ensuring that technological innovation is matched with critical thinking and real-world impact.
“Leading innovation isn’t a single-threaded effort — it demands real-world application, enterprise discipline and deliberate experimentation,” says John Yawney, Chief Analytics Officer and industry partner at Adastra.
“Our partnership with McMaster enables Adastra to scale innovation responsibly while advancing the capabilities of both experts and technology to meet tomorrow’s challenges.”
And with research programs spanning engineering, health sciences, business and other disciplines, explains Gadsden, McMaster is rapidly becoming a national hub for applied, interdisciplinary AI research.
“What’s happening at McMaster is bigger than just individual research programs,” says Gadsden. And it’s not just about coming up with new models or frameworks for AI, either — it’s about using AI as a tool to help accelerate discovery itself.
For Gadsden, the amount of AI work happening on campus indicates a huge appetite to explore the potential for AI in discovery.
“The work we are doing in this field is changing the game for researchers and students alike,” says Carlos Filipe, Associate Dean of Research, Innovation and Partnerships in the Faculty of Engineering.
“It is creating opportunities for us to collaborate across disciplines, tackle real-world challenges, work closely with industry partners and shape the future of technology.”
Opening doors for the next generation
Beyond the applied implications of this work, there’s also a very real stake for current and incoming graduate students.
At a moment when AI is reshaping the global workforce, working directly with a data analytics company to define how intelligent systems are built and trusted places McMaster graduate students at the forefront of a transformative field.
And Gadsden, who also serves as the Associate Chair of Graduate Studies in Mechanical Engineering, is taking note.
“This project is at the intersection of mechanical engineering, computer science and AI,” explains Gadsden. “Because of this, students really get the best of all these worlds.”
Those who join the project gain more than technical expertise — they develop experience working with industry partners, communicating complex ideas to non-academic audiences and seeing their research translated into real-world impact.
“They leave with an advanced degree, specialized training, and a strong professional network,” Gadsden notes. “That combination opens doors across industry and academia.”
This was true for Naseem Alsadi, a former PhD student from Andrew Gadsden’s lab who worked with Adastra during his studies and now works for the company.
“In addition to being research-intensive, my graduate studies at McMaster were accelerated by the fact that I gained valuable industry experience with Adastra at the same time,” says Alsadi.
“This unique experience really helped me start my career in a strong position.”
Students don’t need to come in as experts in everything, Gadsden says. “What matters most is curiosity, motivation and a willingness to learn.”