Getting Comfortable with the Unknown

This summer, I participated in the Learning Directed Operating Systems (LDOS) research project, which explored how machine learning could be used to make operating systems smarter and more efficient. When I first heard about this project, I was drawn to it because it combined two areas I had always been curious about—low-level systems and artificial intelligence. Operating systems are the backbone of everything we do on computers, but they are basically invisible to the user. The idea that I could contribute, even in a small way, to making them more adaptable and intelligent felt both exciting and meaningful. 

While I definitely expanded my technical skills, what stuck with me the most was how much I grew as a person through the research process. One of the biggest challenges I faced this summer was becoming comfortable with uncertainty. Yes, I learned about Linux commands and tracing tools, but I also learned how to deal with the feeling of not knowing what I was doing and to push forward anyway. Almost every day, I found myself needing to do things I had never done before—whether that was setting up benchmarking environments or interpreting long error logs. At first, this was overwhelming. I often felt like I didn’t even know the right question to ask. But slowly, I learned to lean on my resources. Google, ChatGPT, YouTube tutorials, and especially my peers and mentors became essential sources of guidance. Asking for help felt uncomfortable at first, but I realized that research is supposed to involve collaboration. The more I practiced, the more natural it became to admit when I was stuck and to seek out advice. That shift was huge for me, because it reinforced that asking for help is not a weakness, but a skill in itself.