Submit your application by Sunday night to save $1,000. Apply Now

Demystifying Data Science: Navigating a Daily Diet of Data at Grubhub

By Metis  • August 08, 2017


This post was written by George McIntire, a graduate of the Metis Data Science Bootcamp, former freelance journalist, and current Data Science Instructional Associate at General Assembly.
_____

How does the weather affect your food-ordering patterns? Do you eat more takeout in the colder months? Do you order delivery every time a little rain hits the ground?

These are the types of questions Metis bootcamp alumnus Yong Cho has been thinking a lot about lately. As a Data Scientist at Grubhub, he works on figuring out the daily impact of weather on the business.

“Obviously, the food delivery space is one of convenience, so there's significant impact if, say, there's rain during dinner hours in NYC and people don't want to go out to restaurants or grocery stores. This is where some interesting model comes into play, but bottom-line, modeling this out allows us to understand our weather-excluded underlying order growth,” he said. “Weather is something that we can't control...investors are interested to know the “real” growth and performance of the business, excluding order inflation/deflation from the weather. It's a really interesting machine learning problem!"

He’s now been at Grubhub’s headquarters in Chicago for nearly 2 years and has worked on a variety of projects large and small. One of his favorite aspects of the role and the department is that the leadership is very cognizant of keeping things fresh and manageable to avoid burnout.

“We focus on quick deliverables and break long-term projects into smaller chunks, so I'm not stuck doing one aspect of data science for weeks or months on end,” said Cho. “But for me, the most important part is that I'm improving as a data scientist every day at work."

He spends a lot of time on predictive modeling and quick ad-hoc analysis with SQL and pandas, in addition to learning and using Spark and honing his skills in data visualization using Tableau and more. And beyond working on the weather initiative, he’s also navigating a new challenge: learning to deal with codebase handoffs when a data scientist on the team leaves the company.

“Looking at someone else's extensive code can be somewhat overwhelming, so learning how to read thru it and knowing how to better prepare in the future for something similar has been an interesting learning experience,” he said.

Cho is a lover of these sorts of challenges and a lover of data in general. But it was actually his affinity for basketball, chief among other things, that led him to pursue data science in the first place. The popularity of NBA analytics – the rich and abundant data offered by the league – was a major catalyst in his becoming fascinated with the field. He found himself playing around with the data in his free time, digging into stats, trends, and forecasts, before arriving at a decision to quit his day job as a bond trader to give data science a real shot.

“At some point, I realized I'd love to get paid for the kind of data work I enjoy doing. I wanted to develop an in-demand skill set in an exciting up-and-coming field,” he said.

He went through the Metis bootcamp, completing the project-based curriculum, which he says had a significant impact on him securing his current role.

“Whenever talking to a data scientist or hiring company, the impression I got was that companies hiring for data scientists were really, more than anything, interested in what you can actually do,” said Cho. “That means not only doing a good job on your Metis projects, but putting them out there on your blog, on github, for everyone (cough, cough, potential employers) to see. I think spending a good amount of time on the presentation of your project material – my blog definitely helped me get many interviews – was just as important as any model accuracy score.”

But Cho isn’t all work and no play. He gives the following, important advice to any incoming bootcamp student:

“Have fun. In the end, the reason we all joined Metis is because we love this stuff,” he said. “If you're genuinely invested in your subject matter, as well as the skill-set you're learning, it'll show.”

_____

Interested in learning more about the Metis Data Science Bootcamp? Check it out!


Similar Posts

alumni Q5a70ortzpdasyecqexg
Meet Dangaia Sims, Metis Bootcamp Grad and Sr. Data Scientist at IBM

By Emily Wilson • August 23, 2018

In this brief interview, hear Metis grad Dangaia Sims tell the story of her transition from academia (Ph.D., Kinesiology) to industry and how the bootcamp helped her get there.

alumni 3apylgbbr76ekxgom8s4
Q&A with Metis Grad Leon Johnson, Data Scientist and Former U.S. Airman

By Metis • November 05, 2018

We recently hosted a Live Online Ask Me Anything session with Metis bootcamp graduate Leon Johnson, a former U.S. Airmen who was recently hired as a Data Scientist at Viral Launch in Indianapolis. He answered questions from those interested in (or preparing for) the bootcamp, as well anyone else curious about his journey and story. Read the full Q&A here, which includes specific questions about his time in the Air Force and his use of GI Bill funds to attend Metis.

alumni 7gu9puqpq9ex1etqcpw8
Made at Metis: Predicting a Pitch in Baseball, Predicting Mode in a Song

By Emily Wilson • October 18, 2018

Read about two recent final projects created by graduates of our data science bootcamp . See what's possible to create in just 12 weeks, including a way to predict which pitches the Cubs should throw to the Cardinals, and how to predict the mode of a hit single on the Billboard charts.