Mining Data for Evidence in Litigation

Kushal Agrawal (MSAI '23) spent his summer as an applied science intern for Relativity, a software company whose products help users parse large volumes of data to identify key issues during litigation, investigations, and compliance proceedings.  

His job was to research how one of the newest large language models — OpenAI's GPT-4 — could be used to help in e-discovery review, the process by which massive amounts of data are examined to find the important pieces of evidence often buried inside.  

“The opportunity to work on cutting-edge projects involving artificial intelligence was the most exciting part of my work,” said Agrawal, a student in Northwestern Engineering's Master of Science in Artificial Intelligence (MSAI) program. “I gained insights into project management, problem-solving, and the practical challenges of applying AI in real-world scenarios.”  

Artificial intelligence (AI) is turning the once manual, arduous and time-consuming task of plowing through data to find the most miniscule but essential fact or figure into an automated science that can be accomplished in far less time with far less people power. Yet correctly setting up the AI models that do the data diving is of utmost importance when it comes to complicated and high-stakes legal and business affairs.  

That’s where someone like Agrawal comes in.  

Agrawal said he found the perfect balance of autonomy and mentorship while at Relativity, a Chicago-based company founded in 2001 with the mission to help customers organize data, discover the truth, and act on it. 

He said he learned many vital lessons during his internship.  

“One of the most important is the value of experimentation and iteration,” Agrawal said. “Trying different approaches, assessing their outcomes, and adapting based on results is crucial in AI research and development.”  

He also learned how important effective communication and collaboration are to translating technical work into meaningful insights.  

These are skills Agrawal was already becoming more comfortable with prior to his internship thanks to his time in the MSAI program. He said he chose the program to continue his education for a variety of reasons.  

“The emphasis on both theoretical understanding and practical applications, along with the potential for impactful projects, made MSAI a compelling choice,” he said. “The core courses aligned well with my interests and career goals, and the option of pursuing several electives while still being in an accelerated program were also key factors in my decision.”  

That coursework and his successful internship has Agrawal on pace to graduate later this year. He said his goal is to join a company using artificial intelligence to make a positive difference in the world.  

He recommended the MSAI program to prospective students looking to take a similar career path.  

“The core courses are good at providing breadth of learning, so plan your electives in a way to improve your depth in a specialization,” he said. “The experience of working with AI technologies in a professional setting will provide a practical foundation for tackling complex challenges in future research and industry roles.”