Like several other CSL students, Aniket Shirke graduated this month with a master's degree in computer science. Unlike the other students, Shirke’s experience at Illinois involved building computer vision systems for farms and learning how to properly hold a piglet. Read below to see what Shirke had to say about his unique experience and how he will apply the concepts he learned here in his new position at Google.
What are your plans after graduation?
I will be joining Google as a Software Development Engineer for the YouTube Ads team after I graduate on May 15th. There are numerous Ads teams in Google, so specifically, I’ll be joining one of the product teams. A lot of system building goes on behind the scenes, so I’ll be doing end-to-end system design and analysis for different products. Computer Vision and system designing are one of the key concepts I learned on my Illinois journey, and I hope to take these concepts with me to Google.
What was your favorite part of your time at CSL?
One of the most favorite memories of mine is a farm trip we had with my adviser Matt Caesar and Ryan Dilger, Vikram Adve, Narendra Ahuja, Angela Green-Miller, and Isabella Condotta. We visited a commercial farm an hour away from campus to learn how they function and we got to interact with a lot of pigs. That was a very memorable experience! I got to understand how commercial farms work, how rough the farm conditions are, and what camera placements in such scenarios look like. I got a lot of ideas from that trip which I used in my camera deployment for my research thesis.
I also got a chance to take some pictures of baby piglets and I got to learn how to hold a piglet - the trick is to hold them by their stomach, so they don’t feel like they’re falling.
How did you end up working with pigs?
I joined Professor Caesar’s group in the fall of 2019 to do my master’s thesis. At that point, there was a collaborative project under the AIFARMS institute (Artificial intelligence for Future Agricultural Resilience, Management and Sustainability) and we identified that precision livestock management using computer vision was a rapidly developing research field. I was the only graduate student working with the group, so to understand the problem domain, I collaborated with numerous researchers in the animal sciences domain. Since pig behavior is a strong indicator of their health and monitoring them manually is infeasible, I worked on an automated multi-camera pig tracking system that leveraged multiple AI techniques.
The work gave me more exposure to building an end-to-end computer vision system. It was pretty fun because I got to interact with a lot of people from animal sciences, and I also got to decide how my system would be designed from the ground up.
What advice do you have for current students?
Be passionate, get out of your comfort zone, and do interdisciplinary research. I would advise you to find something you are passionate about and then work hard towards it. That’s what drives most of us. In January of 2020, the livestock monitoring thrust of AIFARMS had four professors, with me being the only graduate student. Despite the hurdles introduced due to COVID, we now have more than 20 students in the team, with more professors collaborating, and that’s only possible because all the members of the group are passionate about the vision of the project and the collaboration.
Don’t be afraid to collaborate with researchers in other departments. As computer science students, we are way too comfortable in our own bubble, coding in front of our laptops. But doing interdisciplinary projects will get you out of that comfort zone. Through these projects, you get to broaden your horizons by understanding the research challenges in other departments and how your expertise can help in alleviating some of these challenges.