Students on the Seattle campus traveled to the Silicon Valley campus and vice versa to attend industry panels and participate in a hack-a-thon.
Surya Shivam learned two big lessons over spring break: document your journey and don’t hesitate to reach out for help — takeaways that had nothing to do with traveling to a beach for a relaxing vacation.
Shivam was one of 28 Northeastern University graduate students who spent spring break digging deep into all aspects of AI — from its impact on business to prompt engineering and building their own AI-powered agents.
But AI Immersion week was mostly about skill-building. Especially the hack-a-thon, which challenged students in Silicon Valley to develop an artificial intelligence agent using Agentforce, a new software developed by Salesforce. An AI agent is a system that can process information and take actions autonomously.
“It was all learning, nothing else,” said Harshika Santoshi, a master’s student in data science at the Silicon Valley campus. “For someone who has never touched that application to figure it out, that was a learning experience.”
Santoshi, who stayed in Silicon Valley and hosted Seattle-based graduate students, worked on Shivam’s team designing an AI-integrated invoice fraud detection system. But they didn’t head into the hack-a-thon unprepared.
Before beginning their work, students attended a workshop with Curt Carlson, Northeastern business professor of practice, who shared strategies to hone a product idea before beginning to build it.
“That helped us think about coming up with an idea that is valuable to customers,” said Silicon Valley computer science master’s student Aisha Abdur Rahim. “Because if we are creating something that’s not valuable, then it’s basically a waste of time.”
Working with Agentforce wasn’t easy at first, students said. As representatives from Salesforce roamed the room to answer questions, students began building their projects only to find that some features, like loading images and PDFs, didn’t work. Other students found themselves spending the majority of their time stuck and wished it were easier to back up to an earlier version.
Which is how Surya Shivam learned it is important to always document your work and don’t hesitate to ask for help.
“If our team had done that, our teams would have been able to submit the project well before time,” Shivam said. “When I asked for help, we had a very clear idea of how to proceed.”
The experience came in handy at the Salesforce TDX conference, which students attended after the hack-a-thon. Also attended by professional software developers and IT leaders, the conference included a workshop about Agentforce, which was daunting at first, said Rahim. But that didn’t last long.
“I felt super confident and able to converse with professionals who have been using this tool throughout their work,” she said. “It was a bit intimidating because all of the participants were seasoned professionals.”
Students who spent the week in Seattle learned directly from industry how AI is being used at the corporate and startup levels. Their first day was spent digging into how venture capital works, during a workshop with Seattle-based fund OneSixOne Ventures. Students engaged in discussion with experts who explained how they choose business ideas to support.
This was especially inspiring for Seattle student Jenny Huang, who is getting her master’s in computer science but has a background in finance.
“They gave a very specific presentation about how they evaluate if an AI product is going to be successful or not,” Huang said. “Even though I am a CS student, maybe I will find a career path in the tech industry identifying good tech products and startups.”
Students were encouraged to network at the workshop. Computer science master’s student Kaustubah Eluri said the week was worthwhile after the first day alone.
“I got so many connections on day one itself,” he said. “As a group we learned how important communication is with investors, hiring managers and recruiters across industries.”
During the week in Seattle, students also met with industry practitioners at T-Mobile and Microsoft to understand how AI is influencing their work and product development. Amazon Web Services Skills Center presented a workshop, “Demystifying Generative AI,” to give students an understanding of how artificial intelligence has developed.
“From the theory to the concept in the 1950s, we got the basics about machine learning, deep learning and generative AI,” said Jenny Huang, a computer science master’s student in Seattle. “It was a very good foundation.”
Students also received a deep look at the evolution of prompt engineering for AI-integrated products. Engineering and data science professor Ram Hariharan led a workshop where students explored the different outcomes of basic and more sophisticated prompts.
“We looked at how few-shot prompting can be combined with chain-of-thought methods,” Huang said. “Then we can use a more advanced and comprehensive method called ReAct prompting. It progressed from very basic to advanced methods, which is useful for academics and also daily life and work.”