Exploring LLMs with Accenture Labs

MBAi and MSAI students partnered to examine how generative AI could be used to create brand-aligned marketing content.

Alex Kass is a fellow and principal director at Accenture Labs, where he leads the company's R&D group focused on human-centric computing and human-machine interactions. He's spent two decades at the company focused on digital experiences, and in a field changing so rapidly, he relishes mentoring and learning from students whenever he can.

That is why he was excited to collaborate with students from Northwestern's MBAi Program — a joint-degree program offered between Northwestern's Kellogg School of Management and the McCormick School of Engineering — and Northwestern Engineering's Master of Science in Artificial Intelligence (MSAI) program.

"I jump at every chance I get to shape collaboration that will bring the best of Accenture’s research together with smart students in various fields," Kass said. "I love the cross-disciplinary nature of the MBAi/MSAi programs and the focus on applications of artificial intelligence (AI) to business is right up our alley."

The opportunity was part of the two programs' capstone project, where students collaborate and learn from one another to present solutions to problems facing industry partners. A team of four students — two from MBAi and two from MSAI — worked with Accenture to build on work the company did the previous year examining the use of generative AI and large language models (LLMs) to create brand-aligned marketing content.

"We were excited to roll up our sleeves and get our hands dirty with the inner workings of LLMs," Neil Bhargava (MBAi '23) said, "exploring the pros and cons of different training techniques and applying what we learned throughout our curriculum to a real-world business problem."

Vivek Khetan, a principal researcher at Accenture Labs, was excited by the opportunity to collaborate with the students.

"Working with academic partners is consistently a rewarding experience, and this capstone project with the MBAi and MSAI programs stood out because it offered the chance to collaborate with students from both technical and business backgrounds, mirroring the interdisciplinary approach we value at Accenture Labs," Khetan said. "Two aspects particularly impressed me during our regular interactions: the students' efficiency in collaborating, and their remarkable productivity, especially considering many were tackling such problems for the first time."

The students navigated an ambiguous problem statement, meaning that before they could apply frameworks to the problem, they had to define the scope of what they were solving.

Rahul Saxena (MBAi '23) felt the opportunity was a valuable training exercise to prepare him for opportunities and challenges he will face professionally after graduation. The experience felt less like classwork and more like a simulation of developers and product managers working together at a technology firm, Saxena said.

"For anyone thinking of a career in tech, this is a great exercise," said Saxena, who now works as an associate at consulting firm McKinsey & Company. "You work at the intersection of technology and business on current technical problems and with leading companies and firms attempting to solve them. Plus this is done in a timeboxed manner, which enables you to also learn how to deliver a work output in a specific timeframe and with high quality." 

It was that high quality that particularly impressed Kass. He knew the challenge presented to the students was ambiguous, and he felt they did a good job of finding solutions given the project's time frame.

"I was impressed with how much could be accomplished in a short amount of time, from analysis of the existing technical landscape, to refining the requirements and scoping the effort, running initial examples, and then documenting and presenting the results," Kass said. "It layered a solid foundation for further work. I only wish we could have had time to do some more follow-up work."