Applied Scientist II

The Trade Desk The Trade Desk · Media · Shanghai, China · Data Science

Applied Scientist II role at The Trade Desk, focusing on developing and improving machine learning models for digital advertising. The role involves transforming large datasets into actionable intelligence, contributing to client-facing products, and working closely with product, trading, and engineering teams. Emphasis on shipping product and deriving insights from massive datasets.

What you'd actually do

  1. Develop end-to-end machine learning prototypes that could scale to run in a production environment
  2. Contribute meaningful improvements to existing machine learning models through carefully directed research
  3. Maintain a focus on shipping product, understanding that oftentimes done is better than perfect
  4. Derive actionable insights from massive data sets with minimal support
  5. Provide input into the collection of new data sources and the refinement of existing ones to improve analysis and model development

Skills

Required

  • statistical machine learning
  • optimization techniques
  • deep learning frameworks (Tensorflow, Pytorch, MXNet)
  • big data technologies (Spark, AWS, Hadoop, Hive)
  • SQL
  • Python
  • R
  • Scala
  • Java

Nice to have

  • inference optimization
  • digital advertising
  • recommendation systems
  • recall / ranking

What the JD emphasized

  • shipping product
  • massive data sets
  • recall / ranking is a plus
  • experience recall/matching/retrieval algorithms
  • deep learning frameworks
  • inference optimization is a plus

Other signals

  • Develop end-to-end machine learning prototypes that could scale to run in a production environment
  • Contribute meaningful improvements to existing machine learning models through carefully directed research
  • Maintain a focus on shipping product
  • Derive actionable insights from massive data sets with minimal support
  • Experience in digital advertising and/or recommendation system, relevant experience in recall / ranking is a plus
  • You have experience recall/matching/retrieval algorithms
  • You are familiar with one of the deep learning frameworks (e.g., Tensorflow, Pytorch, MXNet), knowledge of inference optimization is a plus