Remarks at UVA Graduation

William McNamara • August 1, 2022

Speech given at the University of Virginia July 31st, 2022


"Well, here we are.


Let me start by saying how happy I am for everybody here today. It seems like only yesterday our class was meeting for the first time at the Arlington campus. It was in the throws of the pandemic so we were still wearing masks. Which did a good job hiding our terrified faces when our finance professor kicked off the program by cold calling Amanda on the difference between Net Income and Cash Flow.


5 whole modules we’ve survived since then…


In Mod 1, we learned how to ANALYZE, a word I’ve since committed to having tattooed on my body. Eric learned to hedge investments not himself; and unfortunately we all learned that UVA football, well, next year will be our year.


In Mod 2, we learned linear regression, we learned just how many people aren’t paying their mortgages, and we all learned the hard way that trying to compete with Dan Gogue’s machine learning models is futile. Still not convinced he’s not a tenured professor undercover as the Reese’s Guy.


In Mod 3 we…what did we do in Mod 3, does anyone remember? I was face deep in Christmas cookies at the time. What I do know I learned is that contrary to what one might assume, a SQL query will not run just because you yell obscenities at it. And that knowledge alone is apparently enough to qualify you to consult for Hilton. Michelle Tansey enjoyed it so much she decided to go work there!


In Mod 4 we did some genuinely pretty crazy stuff. I don’t know if you all realize how insane it is that we went straight from writing basic SELECT queries in SQL to neural networks, image recognition, and creating deepfakes of Shelly’s dog. I feel like Mod 4 was the MSBA version of a tequila shot. I mean, we learned that Songyuan and Allen are apparently the missing members of Van Halen, and one of our professors almost punched Ben Fishburn in the face. By the way Aidan, just so you know, Tequila is like water that makes you make bad decisions. That’s what our professor’s call a testable hypothesis and anyone who wants to help test that hypothesis, Wes says the job to be done is at Boylan Heights tonight.


But crazy is how the world of data is, everybody here knows it comes at you fast. It’s as scary as it is exciting. Demanding as it is liberating.


I have loved my brief time here at UVA. And by UVA I mean my zoom room provided by UVA. In large part because of the wonderful people I’ve spent it with, but also because of the fascinating things we’ve learned. Things that would have been unthinkable to me 10 years ago.


It’s odd to remember now, but growing up, I did not like math. Not one bit. Until one day in college, I had to take a complex systems course for my research track. It finally connected math to the world around me. And helped me realize that what I was struggling with wasn’t math’s importance, rather a difficulty with the premise that anything in our real world can be simply explained on a single page.


But Complex systems. Complex Systems don’t have right answers. They are sprawling webs of interconnected variables that don’t always behave the way you expect them to, and most of the time you won’t even know why.


This is the philosophy we’ve found woven through every lesson we have learned in the MSBA program over the last year. In Data Science, we must dismiss the idea that we will ever be right. Because our professors taught us that 100% predictive confidence means something is probably wrong with your data, and you should go back and yell some more obscenities your SQL query. Then buy Dustin a beer and ask him to fix it for you.


Side note: my very favorite memory from this program was our Python professor reassuring Laurel that her faltering model was actually the best model because as long as you did the opposite of what her model said you should do, you’d be right.



We know to some extent we will always be wrong. That’s the humbling nature of the world we live in. But every morning we wake up, take stock of what we’ve learned, and do everything we can before the sun goes down to be less wrong than the day before.


I’ve since adopted this as the best framework I’ve found for approaching data science, and all things that are complex, in truth for approaching life.


The world is changing rapidly every second of every day, and no analytical model, no AI, or supercomputer will ever be able to make full sense of it. But each of us can make decisions in our day, big and small, to do what we can to create less wrong. And as in any complex system, we will see emerge an undeniable trend, a greater direction in what seem like little things.


Nowhere will this be more important than in the careers we have ahead of us. Some of the most important ethical questions of our time relate to how we use data.


Where are the boundaries of public and private data? Is it ethical to use machine learning to replicate an individual’s image? Or predict their actions? Or will prerealizing our expectations of each other diminish our capacity to exceed them?


We didn’t learn the right answers to these questions, because there aren’t any. There is only less wrong. Each of us will be co-authors of our future, weather you’re an AI engineer at Tesla or a crash test dummy at Tesla. And my wish for you, my friends, is that we will all meet every tomorrow less wrong than the day before.


I do not believe there is such thing as a perfect data scientist, nor are there perfect humans. But it has been my great privilege to learn from and alongside amazing people, dedicating what little time they have, to be better at both.


Thank you, and congratulations to you all!"


By William McNamara March 19, 2023
Like many music enthusiasts, the most used app on my phone by far is Spotify. One of my favorite features is their daily or weekly curated playlists based on your listening tastes. Spotify users can get as many as six curated ‘Daily Mixes’ of 50 songs, as well as a ‘Discover Weekly’ of 30 songs updated every Monday. That’s more than 2k songs a Spotify user will be recommended in a given week. Assuming an everage of 3 minutes per song, even a dedicated user would find themselves spending more than 15 hours a day to listen to all of that content. That…wouldn’t be healthy. But Spotify’s recommendations are good! And I always feel like I’m losing something when these curated playlists expire before I can enjoy all or even most of the songs they contain. Or at least I did, until I found a way around it. In this articule, I’m going to take you through Spotify’s API and how you can solve this problem with some beginner to intermediate Python skills. Introduction to Spotify’s API Spotify has made several public APIs for developers to interact with their application. Some of the marketed use cases are exploring Spotify’s music catalogue, queuing songs, and creating playlists. You can credential yourself using this documentation guide . I’d walk you through it myself but I don’t work for Spotify and I want to get to the interesting stuff. In the remainder of this article I will be talking leveraging Spotipy , an open source library for python developers to access Spotify’s Web API. NOTE : At the time of writing, Spotipy’s active version was 2.22.1, later versions may not have all of the same functionality available.
By William McNamara March 6, 2022
Online gaming communities need to work harder to close the gap for their female users.
By William McNamara February 15, 2022
Hospitals hold the key to predicting how long a product will be on the shelf.
By William McNamara June 5, 2021
The game you're playing has probably never been played before.
By William McNamara December 21, 2020
Sometimes it's better to build it yourself.
By William McNamara April 6, 2020
Learn what steps you can take to get started as a Salesforce professional.
By William McNamara August 8, 2017
Originally published on govloop.com
By William McNamara August 3, 2017
Originally published on govloop.com
By William McNamara July 25, 2017
Originally published on govloop.com
Show More
Share by: