Welcome to the ninth installment of our blog series “My Path to Google.” These are real stories from Googlers, interns, and alumni highlighting how they got to Google, what their roles are like, and even some tips on how to prepare for interviews.
Today’s post is all about Jasmine Collins. Read on!
Can you tell us a bit about yourself?
I grew up in Pittsburgh, Pennsylvania. I stayed local for college, and went to the University of Pittsburgh, where I double majored in neuroscience and computer science, and minored in chemistry. My dream is to contribute to bridging the gap between computer intelligence and human intelligence. Outside of work, I really enjoy plants and gardening. Also, I love dogs!
What’s your role at Google?
I was on the Google Brain team as a part of the first iteration of the Google Brain Residency Program — a year long training program in deep learning research. For me, it was really excellent to have the opportunity to take a year off between undergraduate and grad school and get some real world work/research experience. Being a Google Brain resident definitely solidified my decision to go to grad school, and helped me bulk up my resume to get into one of the top universities for artificial intelligence/deep learning research — UC Berkeley!
During the residency, I was able to publish my first, first-author paper. In it, we experimentally investigated the tradeoffs across different recurrent neural network architectures in terms of capacity and trainability. The project involved running thousands of optimizations in parallel, over many, many weeks. I really enjoyed this project, as it was something that could only be done with Google-scale infrastructure, but it had findings that could be applicable to the rest of the research community.
What inspires you to come in every day?
My favorite part about coming in every day was the lunch conversations with my regular lunch group. We’d talk about everything from new arXiv papers to crazy startup ideas, debugging code, the validity of Simulation Theory, how long until self-driving cars become a reality, etc. We argued about pretty much every topic in a constructive way, which caused me to think thoroughly and critically about my own beliefs. There was never a dull lunch!
Can you tell us about your decision to enter the process?
I applied for the residency during my senior year in undergrad, mostly as a backup for graduate school. I had previously applied for a summer internship at Google (with dreams of working with the Google Brain team) and made it through interviews, but ultimately never got host-matched. I was pretty thrilled when I was offered the Google Brain Residency position, and didn’t have to think much before deciding to accept it and defer grad school for a year.
How did the recruitment process go for you?
I actually found out about the residency the day before applications closed. On that day, I was going about my business when my adviser asked me to look into TensorFlow and determine whether or not it was worth switching to (at the time we were using Caffe for our neural net training). In doing so, I stumbled across a set of slides for a talk that Jeff Dean gave at a small conference, which talked about recent TensorFlow improvements and also announced the start of a new training program called the Google Brain Residency. It sounded pretty cool, so I quickly repurposed my grad school application personal statement into a cover letter for the job and applied within the hour. Looking back on how competitive the program selection was that year, and how arbitrary it was that I even discovered the program in time, I feel very lucky to have stumbled across that slide deck!
What do you wish you’d known when you started the process?
In retrospect, I wish that I had reached out and spoken to more people about their research ideas during my time at Google. I started out feeling pretty intimidated by all of the great researchers on the team, and I didn’t really realize until closer to the end of my year there that they are almost all very willing and excited to talk about their work with anyone who is willing to listen.
Can you tell us about the resources you used to prepare for your interview or role?
The interviews for the residency consisted of both a research and coding interview. Being a computer science major, and having had a good amount of coding experience, I felt pretty well prepared for the coding interview. I was much less prepared for the research interview, and I honestly don’t think I did a very good job conveying my previous research experience. It’s likely that the only thing that saved me was my enthusiastic proposal for research that I could do at Google, given the resources and mentorship.
My advice for others who are preparing for a research-style interview is to practice by giving a talk to your lab or class about one of your research projects in order to make sure you can give a clear, concise description of your work, and handle any potentially difficult questions.
Do you have any tips you’d like to share with aspiring Googlers?
I think Google really likes people who are passionate about what they do. If you’re passionate about something that is relevant to the position you’re applying for, make sure to express this in your application.
The Google Brain Residency Program is now known as the Google AI Residency Program! Head over to http://g.co/airesidency to find out more about our program, the recent changes, and how to apply.
