Tech Stack

Description

You will design and build end-to-end ML systems to improve catalog quality, including product classification, attribute inference, and deduplication, using billions of items and attributes. You will develop large-scale prediction and generation models, build ML pipelines covering data preprocessing to monitoring, and collaborate with cross-functional teams to integrate ML into core workflows. This role offers a chance to apply cutting-edge techniques like deep learning and representation learning in a hyper-growth e-commerce environment.

Requirements

• 10+ years in software engineering and/or ML engineering
• 4+ years building and operating large-scale ML systems in production
• Deep understanding of ML/AI/deep learning architectures, loss functions, optimization, statistical modeling
• Proficient in Python plus at least one of Scala/Java/PySpark
• Experience with PyTorch/TensorFlow, SQL, Spark, Data Lake/Warehouse, and cloud ML systems (AWS/GCP/Azure)

Responsibilities

• Design and build end-to-end ML systems for catalog quality (classification, attribute inference, normalization, deduplication)
• Develop and optimize large-scale prediction/generation models on billions of product/attribute data
• Architect ML pipelines covering preprocessing, feature engineering, training, evaluation, deployment, monitoring
• Apply deep learning, representation learning, vector modeling, similarity search techniques
• Collaborate with cross-functional teams (Search, Recommendation, Product, Data, Platform) to integrate ML into core workflows