Machine Learning Engineer, Search - Local Services Team

ByteDance ByteDance · Big Tech · Seattle, WA · R&D

Machine Learning Engineer for ByteDance's Local Services team, focusing on enhancing user discovery and ecosystem growth for hospitality, dining, and leisure experiences. The role involves leveraging large-scale ML for search and recommendation systems, aiming to improve personalized relevance, CTR/CVR prediction, and conversion efficiency for billions of users. Responsibilities include designing and implementing full-stack search algorithms, query analysis, ranking, and personalized behavior modeling.

What you'd actually do

  1. Support the local video service business to enhance user discovery of life services such as hospitality, dining, and leisure.
  2. Improve the search experience in local services and promote ecosystem growth.
  3. Utilize large-scale machine learning techniques in search and recommendation scenarios with billions of users to: Improve user shopping experiences & Enhance conversion efficiency.
  4. Design and implement local services search algorithms across the full stack, including: Query analysis, relevance, recall, coarse ranking, fine ranking, and blended ranking. Personalized behavior modeling for relevance computation. CTR (Click-Through Rate) prediction, CVR (Conversion Rate) prediction. Vector recall and value blending.

Skills

Required

  • analytical and problem-solving skills
  • machine learning
  • deep learning
  • NLP
  • coding skills
  • data structures
  • algorithms
  • Linux development environments

Nice to have

  • search algorithms
  • recommendation algorithms
  • advertisement algorithms
  • local life services
  • e-commerce businesses

What the JD emphasized

  • large-scale machine learning
  • search and recommendation scenarios with billions of users
  • CTR (Click-Through Rate) prediction
  • CVR (Conversion Rate) prediction

Other signals

  • large-scale machine learning
  • search and recommendation systems
  • personalized relevance
  • CTR/CVR prediction
  • billions of users