My Grinnell education didn't prepare me directly for the work that I'm doing, but it did give me the tools that I needed to focus my curiosity.
— Hilary Mason
People share things that make them look good.
It turns out it's important to build a product and not just a bunch of data models.
It's easy to find people who can make pretty pictures, and it's easy to find people who can do math. But it's difficult to find people who can do both.
If you find something obscure fascinating, learn as much about it as you can, because there's a good chance it won't be obscure for long.
Don't take advice you see on the Internet too seriously!
Often, people think that individual data is the most valuable thing they can collect. But it's not useful to know what I am doing or where I am, unless you're particularly interested in me, which is weird. But it is very useful to know what a population of people are doing.
You really need to have a lot of empathy for the work you're doing and the people who you're ultimately trying to help, whether that's a business colleague, a boss, or, ultimately, the user of the software you're building.
I hope that, in the future, data is used to empower people and not just for marketing purposes.
It's really fun and challenging to build things that have never been built before.
I'm really curious about people - what their desires and interests are - and bit.ly's data tell me that. It gives me an unprecedented window into human communication and behavior.
To gain the competitive edge, companies must master the ability to innovate.
I'm very lucky to work at bitly, with a data set that allows us to explore human social behavior at the scale of human social behavior.
Deep learning allows you to create predictive models at a level of quality and sophistication that was previously out of reach. And so deep learning also enhances the product function of data science because it can generate new product opportunities.
I think of 'data science' as a flag that was planted at the intersection of several different disciplines that have not always existed in the same place. Statistics, computer science, domain expertise, and what I usually call 'hacking,' though I don't mean the 'evil' kind of hacking.
As your competitors learn more, you'll need to learn, too.
In some ways, chocolate chip cookie recipes are my favorite algorithms. You put a bunch of bad-for-you stuff in a bowl and get a delicious result.
I'm a huge fan of the liberal arts approach of teaching you to think, analyze, and communicate, then sending you out into the world to cause trouble.
Nobody really cares about short links, but people do care about saving and sharing content.
It's a huge competitive advantage to see in real time what's happening with your data.
A lot of people seem to think that data science is just a process of adding up a bunch of data and looking at the results, but that's actually not at all what the process is.
Don't be afraid to push your company in the direction you think is right - you'll either be fired or promoted, and either outcome will end up just fine.
Everyone likes taking their own photos and seeing themselves reflected back.
Cool innovation might happen in startups, but they often lack the resources or the deep expertise in the problems they want to address.
It's rare that you can solve a technology problem with more technology.
When you start at Bitly, you go through this emotional cycle, where first you go, 'Oh my God, this data is amazing.' But then you start looking at it, and you conclude that humanity is completely doomed. Because what people read is cats and Bieber and celebrity gossip and that stuff.
Teaching someone to program is like giving them a superpower.
After graduate school, I joined Johnson & Wales University in Rhode Island as an assistant professor, but I continued to program in addition to teaching and working on research. I built a program that crawled job boards to determine which skills employers value, which helped Johnson & Wales explore ways to improve its curriculum.
A good scientist can understand the current state of a field, pick interesting questions where a success will actually lead to useful new knowledge, and push that field further through their work.
I grew up as a computer scientist, and I've always been fascinated by algorithms.
The job of the data scientist is to ask the right questions.
Companies that rely on licensing a proprietary dataset should expect to be outpaced by competitors using modern data collection techniques and more frequent updates and greater accuracy.
The core advantage of data is that it tells you something about the world that you didn't know before.
The sender and subject line are actually the most important parts of an e-mail because people tend to put more important information in the subject.
Creating things that people want to tie into their identity is the best way to make them spread.
I wish I'd known more about how to build a startup when I was younger.
My job is to analyze our data set to understand it and build products on it. I look at raw data, do the math to clean it up, and build systems to make it easy to understand.
I don't actually, as a general policy, block any sort of cookies. I keep them all turned on, and that's because I'm willing to make the tradeoff that I let companies gather this information about me in return for a better experience.
Value experiences by how much you're learning, and if you're aren't learning, move on.
Data is a tool for enhancing intuition.
I believe there are a lot of inefficiencies in the way technical innovation happens.
I think we need more ambition about using our data to make our lives better.
I decided that since I was trying to teach 'style' of thinking in science and engineering, and 'style' is an art, I should therefore copy the methods of teaching used for the other arts - once the fundamentals have been learned.
I joined bit.ly as chief scientist in October of 2009. The company is a URL-shortener and content-sharing platform; we provide tools for people to share and track links on the Internet.
Technology is giving companies superpowers to compete more intelligently and capture the data behind changing trends, expanding markets, and new opportunities.
Science is the practice of failing repeatedly but learning as you go.
Data science requires having that cultural space to experiment and work on things that might fail.
Data science is the combination of analytics and the development of new algorithms.
Even smaller companies are putting resources behind their analytics teams in the same way they put resources behind engineering and product teams. There are some great tools out there that allow even tiny businesses to use data effectively.
In tech entrepreneurship, even a lot of hack events tend to be overly commercial in that they're designed to produce companies.
