About us:

Trunk Club (www.trunkclub.com) was started to solve a simple problem – shopping for clothes in stores or online just doesn’t work for most guys. It’s overwhelming, inconvenient, and takes way too much time. With Trunk Club, guys discover awesome clothes that are perfect for them without ever having to go shopping. We combine top brands, expert service, and unparalleled convenience to deliver a highly personalized experience that helps guys look their best.

We start by scouring the planet for the very best in men’s apparel, curating an ever-evolving assortment of pieces from our favorite designers. With the best inventory anywhere as our point of departure, a Stylist on our team hand-selects a trunk of items based on each customer’s needs and tastes. They keep what they love and send the rest right back to us, at no cost to them.

Our team is based in Chicago, New York, Dallas, Los Angeles and Washington, DC. We have grown to over 1,000 employees since starting out in December 2009. 

Are you ready to join Trunk Club to carry on the mission? If so, keep reading...

Why is Data Science important at Trunk Club?

At Trunk Club, we develop services that enable the business to make data-driven decisions, from stylists to merchandising.  Every team relies not only on the data we collect, but more importantly in the clever ways that we interact with it to derive information in a clear and presentable manner.  Our stack makes it easy and painless to turn an algorithm into an API and run follow-up A/B tests using a range of multivariate models at your disposal.

Data Science helps drive Trunk Club.  Ever member helps inform the company on the best ways to grow and change.  You will have autonomy to collaborate with every team and be directly connected to decision makers without bureaucracy.

What type of work will the Data Scientist perform?

• Develop and maintain Machine Learning infrastructure that powers our ranking models, member apps, inventory classification, etc.

• Utilize Natural Language Processing to understand text content across products, including reviews and interactions between users, stylists, and products.

• Use Computer Vision (OpenCV) for image content analysis of novel and existing clothing items: classification, quality, attractiveness, similarity, extraction of features for ranking models.

• Identifying suspicious transactions and malicious users for Fraud. Can you beat SaaS fraud platforms?  We’re betting so.

• Determining the optimal inventory levels to help us effectively manage inventory needs while meeting revenue goals.

• Modeling demand and supply situation in real time to optimize the overall shipping efficiency and give valuable insights to the product managers.

• Use a range of clustering techniques to identify latent clusters of customers based on purchases, demographics, styles, etc.

• Research, develop, and implement predictive algorithm using various regression techniques.

• Help maintain and improve our real-time recommendation API.

• Brainstorm and skunkworks new tools to help minimize the risk of experimenting with new algorithms.

• Explore experimental data for overlooked issues, such as low power, false alarms, etc.  Then develop automated processes for that.

We care about the following experience:

• Industry experience building and product ionizing Machine Learning systems from initial analysis to product and actionable insight.

• Ability to explain underlying mathematics of machine learning systems.

• Previous experience writing production quality code and knowledge of weaknesses and common pitfalls of R and Python.

• Knowledge of Bayesian probability and application of Bayesian methodology to common data science problems.

• Strong programming ability (R/Python).  Our team is responsible for delivering actionable insights and enabling data driven decisions.  Candidate must be able to build out an API driven by classical statistics or machine learning when applicable.

• Previous experience with experimental design (Previous research experience is a plus).  Knowledge of when to use the most appropriate analysis (e.g., ANOVA vs. multi-level modeling) and common problems with datasets (e.g., heterogeneity and unbalanced designs).

• Strong SQL experience.  We interact with the BI and Data Warehouse team using MS SQL Server, Postgres, and Redshift.  Candidate must be able to explain complex joins and sub-queries if asked and explore a database on the fly.

• Strong presentation and communication skills.  Our insights drive the business and without the ability to explain those insights accurately but without jargon, we fail.  Previous experience presenting or lecturing to large groups is a plus.

• Able to operate semi-autonomously to accomplish various business needs before a given deadline.  You are given lateral movement and must use that time wisely.

• Highly agreeable and able to take constructive criticism well.

• Master’s in quantitative discipline or equivalent experience. Ph.D. preferred.