Full-Stack Data Scientist, Styling Algorithms

Stitch Fix

About The Position

This role is primarily focused on the recommendation and search engines that power our styling interface. You will be a key contributor to the system that surfaces inventory to our stylists. While you will collaborate closely with the Foundational Models team (who build our base embeddings), you will own the ranking logic, business logic layers, and the deployment of these models into production. You will also prototype and deploy agentic AI workflows to assist stylists in real-time. This role has a direct impact on the business’s bottom line, and throughout the company's history has been a key source of our innovation and competitive differentiation.

Requirements

You have 3+ years of real-world experience in leveraging Data Science to develop customer-oriented solutions with a recent focus on Recommendation Systems and Search
You understand the nuances of A/B testing and can balance algorithmic metrics with business metrics.
You are fluent in SQL and have experience handling large-scale datasets.
You are proficient in building and deploying apps using the Python data ecosystem.
You have a degree in a quantitative field such as Computer Science, Statistics, Physics, Mathematics, or a related field.
You have strong communication skills with business partners.
You are a team player who sees your growth and success intertwined with the growth and success of your peers.
You are inspired to take on new challenges and do not shy away from failure.

Responsibilities

Be a primary technical owner of our human-in-the-loop augmented recommendations.
Work in a team that embraces both LLMs and traditional ML.
Manage experiments to test new features, including communication of results to stakeholders and leadership.
Solve problems related to recommendation ranking, hyperparameter optimization, and assortment generation.
Leverage a decade's worth of rich and unique data about our clients, our merchandise, and their interactions.
Work with our product, product design, and engineering teams to create roadmaps for developing new client products, user features, and infrastructure.