Crash Course

self-proclaimed car enthusiast, Aren Raisinghani C’19, ENG’19, GEN’21, ML’21 recalls frequenting McLaren and Fisker dealerships near his high school in Palo Alto, California.

Now as a data engineer at Tesla’s Fremont office, Raisinghani is thrilled to combine his love for cars with engineering ingenuity. He serves on Tesla’s safety and restraints team, where he focuses on crash safety. “We’re essentially working on safety for all Tesla cars — airbags, seatbelts, seats,” Raisinghani said.

At Penn, he studied computer science and cognitive science in the undergraduate artificial intelligence program and earned his Master of Science degree in systems engineering, which looks at optimization through a mathematical lens. At the University of Pennsylvania Carey Law School, he focused on business and technology. After interning at Tesla, he was happy to join the company permanently.

“It is really rewarding to know the work you do is making the car safer, which impacts people’s lives, and saves lives,” Raisinghani said, adding that he appreciates exposure to different disciplines, like mechanical and software engineering. He spends much of his time crunching lab and real-world data, but the hard work pays off: “The beauty of all this data is you can do exploratory analysis and come up with new initiatives.”

Knowledge of the law in an engineering field provides a different perspective, which I think is very cool. When working in a highly- regulated space, it is helpful to be able to read and interpret legal rules and regulations. It further helps to have a law background when working with legal teams. I’ve always had an interest in the law and engineering entrepreneurship, specifically contracting and business structures. Familiarity with the law can help navigate an entrepreneurial environment.

Normally, day to day, I spend a lot of time coding on my computer. But I do get to go to the crash lab to watch crashes happen. It’s pretty sweet afterward — I get to see the car and check it all out.

A variety of things: programs to gather data from different sources, programs to turn that data into useful information, optimal ways to store and retrieve data, and machine learning and analytical programs to ultimately make our cars safer. Of these, I spend significant time converting data to useful information, a process known as data cleaning. For example, if I’m working with sensor data, the sensors may have limited memory so the type of data they yield might be sequences of cryptic letters and numbers. Then I have to clean and parse this data, so we can use it.

Because we have a ton of data and we’re able to use that data, we come up with new insights based on novel crashes. Maybe there’s a type of crash that isn’t an industry benchmark or required to be tested legally, but you see a significant amount of those crashes in the real world. We identify those gaps and calibrate the safety systems for our cars so they’re safer and continue to be safer. If you’ve already achieved awesome ratings from tests you need to do, well, how do you take that further? We use the data we get, then say, OK, can we make this even more optimal? It ultimately ends up influencing the cars and helping the people inside them.

Note: Crash test photos are captured from the official Tesla YouTube channel.