Data Science Engineer

Grubhub is looking for an innately curious, business-minded, results-oriented Data Scientist to work in our Discovery and Ads teams. We are focused on providing high quality and highly relevant merchant recommendations and ads to users who are exploring restaurants, looking to order groceries, something from the neighborhood pharmacy, as well as those who are searching for something specific across these verticals. As a member of this highly collaborative team you will partner with other data scientists, engineering, and product to deliver new run time models and services not only in Ads, but improve recommendations and search across the Top of the Diner funnel as a whole. You will also be responsible for creating metrics to validate performance of models, proposing new algorithmic approaches to improve our current system, designing A/B tests, and identifying creative solutions to bridge state of the art information retrieval advanced and business and engineering requirements. Additionally you will be driving technology best practices and guiding the evolution of responsive systems. Responsibilities also include but are not limited to creating documentation accessible to both technical and non-technical audiences, mentoring junior data scientists, creating and maintaining automated training jobs, creating metrics dashboards and alerts, advising best algorithmic trade offs to business stakeholders.

Our team practices end to end project ownership and our work focuses heavily on recommendation and classification from content and clickstream. Deep neural networks, transfer learning from pretrained large scale models, classic regressions, fine tuning, and large language models all have a place in our daily lexicon.

The Impact You will Make

• Help the business gain insights from recommendations in search and discovery, with a focus on balancing short and long-term metrics for both ad placements and organic content
• Optimize user experience by balancing compelling ad placements with relevant organic content to maximize overall engagement and drive orders and diner returns
• Assess new algorithms and business policies with a focus on optimizing across competing KPIs like the short and long term impacts of advertisement on our diners creating appropriate metrics that can be leveraged in our a/b tests and monitoring systems
• Collaborate with Product and Engineering teams to understand new product ideas, assess risks and ensure that the necessary data is available
• Discover new and innovative ways to refine what we're doing and question existing assumptions.
• Relentlessly analyze the interplay between ads and organic content, improving overall business performance through data-driven insights and optimization

What You Bring To The Table

• MS/PhD in quantitative discipline (Computer Science, Math, OR, Physics, Engineering, Statistics or other technical field etc) or equivalent experience
• 2+ years doing counterfactual analysis, operational research or causal inference
• 4+ years experience with data analytics, machine learning, or related field
• 2+ years experience in Python machine learning libraries and deep learning frameworks such as PyTorch and TensorFlow
• 2+ years experience scaling model to production
• Experience with large language models like GPT, LLAMA, BERT, or Transformer-based architectures is a plus
• Experience in developing and optimizing ad recommendation systems is a plus
• Experience in data engineering and feature preparation in pyspark, hive,and the python data stack.
• Comfort communicating performance metrics, model details, and features specifications to technical and non-technical audiences
• Ability to keep up with the latest publications and synthesize research into working models
• Deep interest in self-motivated continuous learning