Software Engineer, Machine Learning (All Levels / All Teams)

DoorDash•San Francisco, NY

3d•Hybrid

About The Position

DoorDash is building the world's most reliable on-demand, logistics engine for delivery. We are continuing to grow rapidly and expanding our Engineering offices globally! We are looking for Software Engineers, Machine Learning to build and maintain a large scale 24x7 global infrastructure system that powers DoorDash's 3-sided marketplace of Consumers, Merchants and Dashers. As a Software Engineer, Machine Learning, you’ll be conceptualizing, designing, implementing, and validating algorithmic improvements to the catalog system and our product knowledge graph at the heart of our fast-growing grocery and retail delivery business. You will use our robust data and machine learning infrastructure to implement new ML solutions to make our product knowledge graph accurate, standardized, semantically rich, easily discoverable, and extensible. We’re looking for someone with a command of production-level machine learning and experience with solving end-to-end user problems who enjoys collaborating with multi-disciplinary teams. This role is hybrid with some in-office time expected and will report to an Engineering Manager.

Requirements

M.S., or PhD. in a technical field such as computer science, mathematics, statistics, physics or equivalent.
3+ years ML industry experience with a solid understanding of machine learning algorithms and fundamentals.
Experience building data/feature engineering pipelines at scale using Pyspark and Snowflake SQL.
Experience with building machine learning systems in production by using frameworks such as PyTorch, Keras, lightgbm, scikit-learn, Spark ML, or related.

Nice To Haves

Online advertising
Search relevance & ranking
Recommendation system

Responsibilities

Develop advanced machine learning models to improve ads efficiency and quality.
Design and build optimization algorithms for budget pacing and automated bidding to achieve various advertising goals.
Establish a data-driven framework to understand how the bid density and market competitiveness would affect advertising value and platform revenue.
Develop new data solutions (eg. embeddings and consumer profiles) to target the relevant audience.
Be responsible for the end-to-end ML lifecycle, including ideation, offline model training, online shadowing/deployment, experimentation, and post-launch monitoring/measurement.
Build and extend the current data/ML infrastructure to empower Ads data applications including data analysis, ML modeling, and experimentation.
Scale our systems and services to fuel the growth of our business.