You will join a team that devises online, offline, or hybrid machine learning/data mining/statistical models for enabling sophisticated analysis, accurate recognition, and quantitative evaluation on over 100 million patent documents and other related data, e.g. prosecution, litigation, technical standards from worldwide. The key responsibility for this role include:
- Tackle challenging problems and collaborate with experts from diverse fields and cultures including US, Japan, Taiwan, and China patent attorneys and agents.
- Design, implement and validate data-driven models in rigorous ways, or effectively leverage state-of-the-art techniques to rank the global patents and analyze the user behavior data.
- Build time-efficient machine learning, data mining, statistical and feature engineering pipelines on an on-premise cluster with hundreds of cores and GPU servers.