MSAI Trio Turn Practicum Project Into Full-Time Jobs

When Bodhisatta Maiti used to think of Home Depot, the first thing that came to mind was it was a retail giant in the United States.

That's why Maiti jumped at the opportunity to work on a project for the company as part of his Practicum in Intelligent Information Systems, a required course for all students in Northwestern Engineering's Master of Science in Artificial Intelligence (MSAI) program.

Today, he doesn't just call the store a retail giant. He also calls it his employer.

Maiti (MSAI '22), Akshay Kumar (MSAI '22), and Dhruv Narayan (MSAI '22) all worked on the practicum project, which involved using machine learning and computer vision techniques to identify the types and number of products customers had with them during check out. The three said Home Depot was happy with their work, and they were given the opportunity to expand on it as summer interns. After that, the three became part-time workers with the company while they finished their MSAI capstone project. 

Today, all three of them work for Home Depot.

Maiti is a staff machine learning engineer for the company, while Kumar and Narayan are senior machine learning engineers.

"All three of us were interested in the Home Depot project because this was a computer vision project and we'd be doing this at the industrial scale," Maiti said. "We were certainly excited to see how we can apply [artificial intelligence] techniques in the retail domain."

During the practicum, the trio helped develop a system that would allow Home Depot to track inventory and make sure no items were lost or inadvertently not scanned. As part of their summer internship, the three focused on image processing to compare different image quality metrics and develop an autonomous system that would separate better-quality photos. They also focused on enhancing object-tracking techniques.

"They really liked the work we did as part of practicum, and that got converted into our internship," Kumar said. "They again liked our work during the internship, so that got converted into part-time work. From there, they wanted us there and offered us full-time jobs."

MSAI deputy director Mohammed Alam has watched Home Depot become a visionary leader when it comes to AI. The company brings that innovative perspective to its partnership with MSAI, he said.

"Our colleagues at Home Depot have a deep appreciation for AI and a strong sense of how they can leverage the technology to elevate the efficacy of the company's operations and improve the customer experience," Alam said. "They've proposed advanced ideas for using AI in their business, and our students have absolutely loved working with the company on identifying viable solutions."

One of the initial appeals for Kumar, Maiti, and Narayan to work with Home Depot was they would work with real data they would not have access to otherwise.

"You're not going to be able to work on these types of problems if you're not working as part of a company," Kumar said. "If you're trying to do something on your own, you won't have access to this kind of data."

Kumar, Maiti, and Narayan all credited MSAI with helping prepare them to excel in their work with Home Depot. Beyond the core fundamentals they learned in the classroom, the three said they learned valuable lessons from the industry leaders who spoke as part of the program's Frameworks Speaker Series. They also appreciated the opportunity to work and interact with programs and schools across the Northwestern campus, including MBAi, a joint-degree program between the Kellogg School of Management and Northwestern Engineering that brings AI into the traditional MBA lessons. 

MBAi and MSAI students collaborated on their capstone projects.  

Though they are still in the early months of their full-time work with Home Depot, the three said they've recognized the uniqueness of their opportunity. They also understand how the solutions they developed and continue to work on can ultimately make a difference for Home Depot.

"We knew our work could have a high impact," Maiti said. "The solutions we're developing will benefit the company."