University IT
Pritish Yuvraj
AI Frontiers:
Dialogues with Tech Pioneers Podcast
Guest: Pritish Yuvraj
Transcript
Any views, discussions, or opinions expressed in this podcast do not represent in any way the opinions or positions of Stanford University, its staff, or its employees.
Don Cameron:
Welcome to AI Frontiers, Dialogue with Tech Pioneers, the podcast where we explore the future of artificial intelligence with leading experts and innovators in the field. I'm Don Cameron from the Stanford University technology training team. Today, I'm joined by Pritish Yuvraj, a machine learning engineer at Meta. Pritish has made incredible contributions to AI, focusing on advancing multilingual capabilities and ethical applications. His work spans child safety, content integrity, and well-being projects.
Pritish, welcome to AI Frontiers! We're thrilled to have you.
Pritish Yuvraj:
Thank you so much, Don. It's an honor to be part of your podcast, and thank you for having me.
Don Cameron:
Thank you! We really appreciate you taking the time to spend with us today. I want to get started by learning more about yourself. What inspired you to pursue a career in machine learning and natural language processing, and how did that journey lead you to your current position at Meta?
Pritish Yuvraj:
That's a great question! So, I was in India doing my undergraduate studies in computer science when I came across a course from Stanford on machine learning. I took the course on Coursera and realized the potential of AI. I could see that the future is about the Internet and AI because there is so much new data available that can be used to derive meaningful insights.
We are blessed with so much computational power today. When you combine data with compute, I think this is the perfect time to enter the field of machine learning. That’s what drew me into AI.
Don Cameron:
Great! And how did your journey lead to your current position?
Pritish Yuvraj:
My journey has been a blessing. I don't think there could have been a better time to be in this field. AI and machine learning professionals now have the ability to impact almost every domain—healthcare, construction, civil engineering, and policy-making. It’s completely unprecedented. I’m enjoying this space and all the opportunities I’m getting thanks to this field.
Don Cameron:
You earned your master's degree at the University of Massachusetts in Amherst, focusing on machine reading, question answering, and monitoring health using NLP techniques. What challenges or breakthroughs stand out for you from your time there?
Pritish Yuvraj:
During my master's, I focused on two main projects. One was food monitoring using natural language processing. The idea was to use NLP and machine learning to help people understand what they’re eating and how it affects their long-term health, especially if they have existing diseases.
One key innovation from this project was learning about word sense disambiguation. For example, the word "honey" can mean different things in different contexts. Back in 2017-2018, we didn’t have large language models, so we had to use additional NLP techniques to improve word sense disambiguation, which is an NP-hard problem. I’m very happy we were able to make progress on that project, and I think now is the right time to continue it, given the advancements in technology.
Don Cameron:
In your current role at Meta, what do you find most rewarding about the work you're doing?
Pritish Yuvraj:
The most rewarding part of being at Meta is the opportunity to work with cutting-edge technology. We believe this is what the world will look like in the next five years, and it’s going to transform everything. We have the resources and opportunities, which is very rewarding. Unfortunately, I’ve seen that the research sector, especially universities, often lacks the necessary GPUs, which are crucial for advancing large language models and even for achieving AGI at some point.
