Senior Data Scientist, Personalization & Recommendations (Remote)

Shoptalk; United States; Hybrid; USD 160,000 - 190,000; Full time

Benefits offered by this job

Medical

Dental

Vision

PTO

401K matching

Job summary

Shoptalk is seeking a Senior Data Scientist to own the personalization engine, from data signals to production models. You will design, build, and evaluate models that capture user taste across brand, category, color, price, and fit, while operating in a hands-on, full-cycle role.

You'll work within an MLOps-enabled platform, focusing on cold-start strategies as the catalog scales toward tens of millions of products. NYC-based hybrid with three days in the office.

Qualifications

• Master's degree or higher in Computer Science, Statistics, Machine Learning, Applied Mathematics, or a related field.
• Experience designing, training, and deploying embedding models and vector retrieval for product/catalog similarity.
• Strong data science fundamentals: statistics, experimental design, evaluation methodology.
• Proficient Python, SQL and production ML tooling; able to ship production-quality code.

Responsibilities

• Own the design and development of core recommendation models for the personalized feed.
• Define offline and online experiments; interpret results to inform product decisions.
• Turn signals and catalog attributes into meaningful features using SQL and ML techniques.
• Take models from prototype to production within the existing MLOps stack.

Skills

Python

PyTorch

TensorFlow

SQL

BigQuery

NLP/Computer Vision

MLOps

Recommendations

Education

Master's degree or higher in CS/Statistics/ML

Tools

Milvus

Pinecone

Vertex AI Vector Search

BigQuery

Job description

Shoptalk is seeking a Senior Data Scientist to own the personalization engine, from data signals to production models. You will design, build, and evaluate models that capture user taste across brand, category, color, price, and fit, while operating in a hands-on, full-cycle role.

You'll work within an MLOps-enabled platform, focusing on cold-start strategies as the catalog scales toward tens of millions of products. NYC-based hybrid with three days in the office.