The Search Engine organization designs, builds, and operates highly performant, fault-tolerant and scalable distributed search engine used by millions of Amazon customers every day. It is built with Apache Lucene at its core.
We are evolving our search engine to better understand natural language queries. This involves leveraging LLMs and other Deep Learning models in different parts of the search stack to improve the search experience.
Ideal candidates are expected to have a background in distributed systems and machine learning infrastructure. Strong candidates will have knowledge of LLM fundamentals and some experience in GPU programming.
Key job responsibilities
- Understand and improve state-of-the-art in Vector Search
- Optimize inference for semantic matching and ranking models
- Improve search engine efficiency to reduce infrastructure cost
- Contribute to open source Lucene and collaborate with internal and external technology leaders.
- Participate in oncall and raise the bar on operational excellence
