Patrick Hayes

Bachelor of Mathematics, University of Waterloo, Computer Science/Pure Mathematics Double Honours, 2012

• Graduated with Distinction; Cumulative average: 90%; Major average: 94%
• Dean's Honours List, University of Waterloo, 2007–2012
• Excellent Academic Standing, University of Waterloo, 2007–2012
• St. Jerome's University President's Scholarship, University of Waterloo, 2007–2009

Relevant Coursework

• Machine Learning; Artificial Intelligence; Algorithm Design; Database Systems Implementation; Principles of Programming Languages; Operating Systems; Theory of Computation
• Readings in Pure Mathematics; Algebraic Number Theory; Measure Theory and Fourier Analysis; Real Analysis; Complex Analysis; Ring/Group Theory; Elementary Differential Geometry
• Communications in Math and Computer Science, Macroeconomics

Senior Staff Software Engineer, Airtable, 2023–present

• Tech Lead for Airtable's Developer Infrastructure team
• Responsible for driving down CI times and costs, saving thousands of eng-hours / month
• Built tooling to accelerate local workflows for development, typechecking, and testing

Senior Staff Software Engineer, YouTube, 2021–2023

• "Uber Tech Lead" for YouTube's Developer Experience team
• Architect and technical mentor for frontend, backend, and infrastructure teams
• Developed test infrastructure for thousands of engineers at YouTube

Cofounder & CTO, SigOpt, 2014–2021

• Founder and architect at SigOpt, which was acquired by Intel in Oct 2020
• Inventor of SigOpt's patented distributed optimization platform for machine learning models
• Full-stack engineer with experience managing frontend, backend, and research staff
• Y Combinator alumni, Winter 2015

Senior Software Engineer, Foursquare, 2012–2014

• Developed passive awareness notifications to remind users of the best content nearby
• Built the tips timeline, for discovering tips from people and tastes you follow
• Created Foursquare expertise, for recognizing and rewarding city and neighborhood experts

Internships, 2008–2012

• Bloomberg, Financial Software Developer, 2011
• Wish, Research Engineer, 2011
• Facebook, Software Engineer, 2010
• Sybase, User Interface Developer, 2009
• BlackBerry, Automation Tools Developer, 2008

Young Alumni Achievement Award, University of Waterloo, Faculty of Mathematics, 2017

https://uwaterloo.ca/math/alumni-friends/faculty-mathematics-alumni-awards/gallery-past-recipients/recipients-faculty-mathematics-alumni-achievement-medal

Scaling TypeScript to thousands of projects. Michael Mitchell, Patrick Hayes. Airtable Eng Blog, 2024

https://medium.com/airtable-eng/a-leap-in-the-evolution-of-airtables-codebase-scaling-typescript-to-thousands-of-projects-734326c3a553

A Strategy for Ranking Optimization Methods using Multiple Criteria. Ian Dewancker, Michael McCourt, Scott Clark, Patrick Hayes, Alexandra Johnson, George Ke. Proceedings of the Workshop on Automatic Machine Learning, 2016

http://proceedings.mlr.press/v64/dewancker_strategy_2016.pdf

Evaluation System for a Bayesian Optimization Service. Ian Dewancker, Michael McCourt, Scott Clark, Patrick Hayes, Alexandra Johnson, George Ke. ML Systems Workshop at ICML, 2016

http://arxiv.org/abs/1605.06170

A Stratified Analysis of Bayesian Optimization Methods. Ian Dewancker, Michael McCourt, Scott Clark, Patrick Hayes, Alexandra Johnson, George Ke. AutoML Workshop at ICML, 2016

http://arxiv.org/abs/1603.09441

Systems and methods implementing an intelligent optimization platform. Patrick Hayes, Michael McCourt, Alexandra Johnson, George Ke, Scott Clark. SigOpt, 2019

https://patents.google.com/patent/US10217061B2

Systems and methods for implementing an intelligent application program interface for an intelligent optimization platform. Alexandra Johnson, Patrick Hayes, Scott Clark. SigOpt, 2019

https://patents.google.com/patent/US10282237B1

Systems and methods for implementing an intelligent machine learning optimization platform for multiple tuning criteria. Bolong Cheng, Olivia Kim, Michael McCourt, Patrick Hayes, Scott Clark. SigOpt, 2020

https://patents.google.com/patent/US10528891B1/en

Systems and methods for an accelerated tuning of hyperparameters of a model using a machine learning-based tuning service. Michael McCourt, Ben Hsu, Patrick Hayes, Scott Clark. SigOpt, 2020

https://patents.google.com/patent/US20200019888A1/en

Accelerating Model Development by Reducing Operational Barriers, NVIDIA GTC, 2019

https://www.youtube.com/watch?v=OMYF9fvK5KQ

Insights for Deep Learning Optimization, Open Data Science Conference, 2018

https://www.youtube.com/watch?v=rj3n2RukuDM

Lessons Learned for Founding a Startup, University of Waterloo, 2018

https://www.youtube.com/watch?v=KjKV6aiwbOk

Optimizing Everything with Python, Talk Python To Me Podcast, 2016

https://talkpython.fm/episodes/show/51/sigopt-optimizing-everything-with-python

Wiktionary Mobile, Wikimania, July 2012

http://www.patrickhay.es/talks/wikimania2012.pdf

Wiktionary Mobile App, Wikimedia Foundation

• Core developer of the official Wiktionary mobile app for Android/iOS
• Worked on a remote team with 3 undergraduate students from schools across Canada

Federation Orientation Committee Member, University of Waterloo, 2010

• Worked with a team of three students to plan Math Orientation Week 2010
• Organized and ran events for 1500 students spanning one week in September
• Contributed over 500 hours of volunteer involvement to planning Orientation Week

Orientation Week Head Leader, University of Waterloo, 2008–2011

• Orientation week Lead for over 200 first year students
• Directed a group of ten other orientation leaders

Undergraduate Teaching Assistant, University of Waterloo, 2009

• Tutored undergraduate students in Calculus
• Marked weekly assignments

Mathematics Society First-Year-At-Large, University of Waterloo, 2007

• Elected by the first-year Math student body to represent them in student government