Principal Full-Stack Data Scientist - Foundational Models

Stitch Fix

4d•$200,000 - $237,000

About The Position

Stitch Fix is redefining retail by blending the art of fashion with the science of machine learning. The Foundational Models team builds and evolves the machine learning systems that power personalization and recommendations across Stitch Fix. The team owns foundational capabilities—including client and item representations, recommendation and retrieval systems, and core models such as the Client Time Series Model (CTSM)—that are leveraged across multiple client experiences and downstream applications. We foster a culture of ownership where Data Scientists manage the full lifecycle of their work. As a Senior Full-Stack Data Scientist, you will work across the full machine learning lifecycle: identifying opportunities, developing and evaluating new modeling approaches, running large-scale experiments, and deploying and operating models in production. You will tackle technically challenging problems in personalization and recommendation systems, working with large-scale behavioral, transactional, textual, and visual data. This is a highly collaborative and hands-on role. You will partner with Data Scientists, ML Engineers, Product, and other technical teams to translate research and emerging ML techniques into reliable systems that create measurable client and business impact.

Requirements

Bachelor’s Degree in a quantitative field such as Computer Science, Statistics, Physics, Mathematics, or a related field required. Master’s Degree or PhD preferred.
8+ years of experience in design and deployment of machine learning solutions, ideally in personalization, such as recommendation systems, representation learning, or search.
Strong ability to architect technical solutions and write production-grade code in Python.
Ability to independently drive ambiguous machine learning problems from initial exploration and prototyping through production deployment, monitoring, iteration, and measurable impact.
Experience working with large-scale datasets using SQL and distributed data-processing technologies such as Spark.
Experience with modern deep-learning frameworks such as PyTorch or TensorFlow.
Strong understanding of model evaluation and experimentation, including offline evaluation, A/B testing, and translating model improvements into measurable product or business outcomes.
Ability to reason about the tradeoffs involved in production ML systems, including model quality, latency, scalability, reliability, and computational cost.
Strong communication and collaboration skills, with the ability to influence technical decisions and communicate complex concepts effectively to both technical and non-technical stakeholders.
Intellectual curiosity and a demonstrated ability to learn and apply new machine learning techniques to practical problems.

Nice To Haves

Master’s Degree or PhD preferred.

Responsibilities

Design, develop, evaluate, and productionize machine learning models that improve Stitch Fix's personalization and recommendation systems.
Advance foundational modeling capabilities, including client and item representations, embeddings, retrieval, ranking, recommendation models, and assortment generation.
Explore and apply LLMs and machine learning techniques—including deep learning, representation learning, multimodal modeling, and generative approaches—where they can meaningfully improve our systems.
Own work across the full ML lifecycle, from problem formulation and data exploration through modeling, experimentation, deployment, monitoring, and iteration.
Design rigorous offline evaluations and online experiments to understand model performance, measure client and business impact, and communicate results to stakeholders and leadership.
Work with large-scale behavioral and product datasets using Python, SQL, and distributed data-processing tools.
Build production-quality ML solutions with attention to scalability, reliability, latency, observability, and cost.
Collaborate with Product, Engineering, and other Data Science teams to understand downstream needs and translate them into reusable foundational ML capabilities.
Contribute to the technical direction of the Foundational Models team through design discussions, code reviews, research, prototyping, and development of engineering and modeling best practices.
Communicate complex technical concepts, modeling approaches, and tradeoffs, and recommendations clearly to both technical and non-technical partners.
Mentor and collaborate with other Data Scientists and engineers, helping raise the technical bar across the team.