Pinterest Search currently operates over a very large scale of hundreds of millions of queries and billions of documents, and our next step is to dramatically improve our search results by customizing them to a particular user. We plan on using significant personalization to build a truly unique search product. To achieve this, we'll be leveraging this huge amount of data by building an efficient and scalable full search stack capable of combining traditional information retrieval technologies with efficient user personalization models.
- Work on improving next generation Search across Pinterest and enabling result set personalization
- Partner cross-functionally to enhance the performance and reliability of the core search system as well as enhance our experimental new hybrid search
- Develop a very large scale system utilizing clusters of machines to implement a custom full stack search strategy involving retrieval, lightweight scoring, full scoring and result set analysis as well as query understanding
- MS or PhD in computer science or closely related field
- 5+ years of relevant industry experience
- Experience in the design, implementation, and delivery of sophisticated production infrastructure systems
- Fluent in C/C++, Java and/or Python