Amazon Legal Technologies is an innovative and high-energy team that supports Amazon’s legal operations worldwide. We focus on creating easy to use, highly secure, ingenious solutions that allow our business to efficiently scale for the future. We are looking for a smart, creative, and results-oriented Software Engineer who is excited by unique challenges and is passionate about using the latest technologies to solve customers’ needs. This is an excellent opportunity for a developer who is looking for a growth position with a high level of ownership in a small, growing team.
Think big! Own, design and create innovative new solutions that allow the Legal department to scale and improve customer service without increasing headcount. As a key member of our team, you will have the opportunity to work directly with your customers, owning major deliverables across all aspects of development: scoping, design, implementation, and testing. Successful candidates can multi-task, quickly adapt to new development environments, learn new systems, create reliable/maintainable code, and find creative and scalable solutions to difficult problems. Candidates applying to this position should be enthusiastic about sharing their previous experiences and lessons with our team to help us collectively succeed.
More about this project - while we are building smart applications that makes life easy for our customers, we have large number of documents we don’t have much information about. It blocks our customers to easily and efficiently discover information and action items. Our small, result oriented data lens team will work on classification of documents, extracting entities and action items from them using Text Mining. We will use Natural Language Processing and Machine Learning to mine information from unstructured data spread across millions of documents. Few key problems -
· Recovering different type of documents from document stores and make them available to related applications.
· Extract metadata and entities from documents to make them searchable and tag them in related applications.
· Extract key information, clauses, and action items from different documents based on their type.
· Handle security aspect of all documents and make sure same is propagated when updating other applications.
· Set up data pipeline to enable our customers to train extraction model by providing training data, test model and launch them to be used in production for future needs.
As an engineer in this project, you will work on Machine Learning, Natural Language Processing and Unstructured Information Management pipelines to build new models to solve complex Text Mining problems. Apart from training and building machine learning models, you will also design a self sustainable system that enable our users to use it and train with different scenarios. Metadata extraction models will empower us to build knowledge base for Amazon legal which will be used by another product ‘Legal Central - Virtual Assistant For Your Legal Work’. This platform will surface Legal knowledge via Chatbot and contextual search.
· Research and evaluate different NLP and Machine Learning libraries, build models to cater Legal needs and enable Legal applications to use the knowledge.
· Establish scalable, efficient, automated knowledge discovery via Text Mining platform by building data processing pipeline to train and evaluate models.
· Build Full Stack application to market AI solutions.
· Mentor and develop junior engineers in the team.