United States
Hello, my name is Kevin Quinn and I'm from Alberta, Canada. I've graduated with a B.Sc and M.Sc in Computing Science at the University of Alberta . I've worked with a variety of technologies and topics over the course of my education. My primary interests are focused on Machine Learning and Natural Language Processing.
Working on recommendation problems.
Joined the Facebook Assistant team leading the efforts for our Smart Replies product. Needed to work with cross-functional partners to build the user experience as well as train the underlying model. Launched the Smart Replies model on both Oculus Messenger and Oculus Chat with a CTR of ~10%. Took the existing LightConv Smart Replies model and both made it generative and multi-task. Fine tuned LLaMA for classification purposes, led effort to reduce annotation costs.
Working with the Alexa A.I. group. Applied state-of-the-art machine learning techniques for improving Alexa's response through query rewriting. Implemented and maintained a model which predicted the customer's experience using Alexa which was used to further refine the model as well as for A/B testing.
Worked with the Consumer Payments organization to provide credit card recommendations for customers. The various credit card advertisements (Discover, Chase, etc) were placed on the Amazon retail site and my work helped rank them to prioritize click-through or sign-up. Built hundreds of logistic regression models, one for each product, using millions of impressions.
Taught various courses such as the web development course and introduction to C/Python.