We research and develop end-to-end question answering techniques, deployed to Alexa world-wide.

A day in the life
Research and development of models for question answering, fact extraction, and reasoning. Publish your findings and innovations in papers.

About the hiring group
We research and develop end-to-end question answering techniques, deployed to Alexa world-wide.

Job responsibilities
We have an exciting position for a NLP/ML scientist to join Alexa AI. Our team makes Alexa smarter by delivering an end-to-end natural language question answering (QA) technology. We build advanced QA models based on constructing a high precision large scale knowledge graph from multiple sources (e.g. facts extraction from text at Internet-scale; linking and aligning open and proprietary knowledge-bases); developing natural language understanding models; and generating natural language responses based on query results on our knowledge graph.

To achieve our ambition we need to develop methods that lie beyond the cutting edge academic and industrial research of today and as a scientist, you will bring academic and/or industrial practical experience and create novel solutions to complex problems at massive scale. We are particularly interested in problems of fact extraction, entity linking, natural language understanding, semantic parsing, natural language generation, masked language models, cross-lingual NLP models, and weakly supervised methods of learning (self-supervised, transfer learning, semi-supervised, curriculum learning).

As a research led team, we have been publishing and contributing to the scientific community and you can find some of our recent work at: Rongali et al, "Don't Parse, Generate! A Sequence to Sequence Architecture for Task-Oriented Semantic Parsing", https://arxiv.org/abs/2001.11458; Harkous et al, "Have Your Text and Use It Too! End-to-End Neural Data-to-Text Generation with Semantic Fidelity", https://arxiv.org/abs/2004.06577; Sen et al, "What do Models Learn from Question Answering Datasets?", https://arxiv.org/abs/2004.03490; Thorne et al, FEVER: a large-scale dataset for fact extraction and verification, https://arxiv.org/abs/1803.05355.

BASIC QUALIFICATIONS

· PhD Degree in Computer Science, Machine Learning, Computational Linguistics, Natural Language Processing, Applied Mathematics or a related field
· Strong academic record of refereed publications in top tier conferences or journals
· Experience of building ML and NLP models in Deep Learning frameworks in Python

PREFERRED QUALIFICATIONS

· Post PhD research or industry experience
· Expertise in: fact extraction, entity linking, question answering, semantic parsing, natural language generation,