“My story might be different from most other Computer Science Ph.D. students because I joined Rice as a student in the professional master’s program,” said Yun Yu, a software engineer at Facebook.
“In my first semester, I took Automata Theory with Luay Nakhleh. As a teacher, Luay is the most amazing one I’ve ever met. He is extremely talented and he made what could have been a boring theory class very interesting for me. I ended up doing very well in that class, then one day Luay asked me if I wanted to do research with him.”
Yu said that Nakhleh gave her a brief introduction about the focus of his research and she quickly agreed to join his research group. “I said ‘yes’ and switched to the Ph.D. program. That’s the best decision I’ve ever made, one which I never regretted for even one minute in my next seven years in Rice.”
It was a big switch for Yu, who had not thought much about research in her undergraduate years. “It was the type of thing that felt really far away from where I was, but Luay convinced me. You will never know if you will like research unless you try it. For me, I ended up loving it — so much that I spent two more years after graduation working as a postdoc with Luay.”
She said the role of a postdoc allowed her the opportunity to determine whether she want to pursue a career in academia or industry.
“I only loved the research aspect of academia,” said Yu. “Being a professor means you also need to do things like teaching. So I chose industry, and Facebook has been a very good fit for me. After 10 years in CS, I know I love coding.”
As much as she admired her adviser, she did not follow his counsel about choosing a career. “Luay always said it was not the best use of one’s training as a researcher to switch from solving problems no one else can solve to becoming a coding machine. Well, it’s true to some extent. The difficulty you face as a researcher is that sometimes you get stuck. You don’t know how to proceed, can’t find a solution. Working as a software engineer in industry, nothing is that hard. The only problem is time: there is not enough, compared to the work you have.”
She credited Nakhleh for proposing solvable research problems. “He’s very knowledgeable about the research we are doing. If he gives us a project to do, it’s usually doable. Sometimes researchers work for months to solve a problem and find it is not doable, but that never happened to me.”
After giving a brief overview of an idea and discussing it with Yu, Nakhleh expected her to explore parameters that would lead to a practical solution. “But the general direction he gives is always correct,” said Yu. “Plus, his door is always open. We could walk in at any time and ask him questions, even silly questions.”
In fact, Yu said Nakhleh encourages his graduate students to ask him silly questions. Yu said, “Once he even said, ‘You’ll spend too much time trying to filter out the information you are seeking if you google it. Just stop by and ask me, it will save you time.’ That felt amazing – that he would take time out of his own busy schedule in order to save us time searching for something that might turn out to be insignificant.”
Yu said she enjoyed her years at Rice so much she would have gladly remained there forever, but family considerations prompted her move into industry where she has discovered a very satisfying career at Facebook.
“Remember, I love coding, so I am very happy here,” she said. “But other Ph.D. students should recognize that a good day as an employee and a good day as a researcher are very different.
“As a postdoc, a good day at work might have been when I read a good paper, had an inspiring brainstorming session with other students, or had a really good discussion with my adviser about potential projects. It was easy to have a good day even if I didn’t code anything. But as an employee in the tech industry, if I don’t finish enough of the coding work I’ve set for myself, then I may feel dissatisfied because I know more work will be coming in tomorrow.”
In addition to the difference in how she measures a good day, Yu also noted a difference in project ownership. “In industry, in most of the cases you may only contribute to part of the project, and your progress can be impacted by colleagues working on other parts of the system. As a postdoc, you own the whole project; how the project goes depends only on you.”
Abandoning a project is also perceived differently in academia. “Your industry team may have put in a lot of effort on a project only to have it cut for other reasons. You may not have anything to show for your efforts. As an academic researcher, if you have good ideas and make good progress, you can still publish a good paper even if the project doesn’t end up as a prototype or in trials.”
Her advice for Ph.D. students is to develop a close relationship with their adviser. “Sometimes you may not make progress on your project for weeks or even months. That’s normal because you are trying to solve some problem that no one in the world has a solution for so far.
“When students get frustrated, it’s very important to be open with your adviser. Advisers can give you helpful advice on how to proceed, or you can ask for another project to work on in parallel. Sometimes stepping back for a while can help you think out of the box. Also, talking to other students about their research can be very helpful. What they are working on might not be related to your area, but their ideas may still inspire you.”
Although Yu was surprised to find herself pursuing a Ph.D., she discovered her unexpected journey to be very rewarding. “I know 99% of people say getting a Ph.D. is hard, but to be honest, that was not the case for me. Luay is an amazing adviser and he made the process much easier for me. I’m very honored to have worked for him for seven years.”