Job Description

DESCRIPTION

Job summary
Customer addresses, Geospatial information and Road-network play a crucial role in Amazon Logistics' Delivery Planning systems. We own exciting science problems in the areas of Address Normalization, Geocode learning, Maps learning, Time estimations including route-time, delivery-time, transit-time predictions which are key inputs in delivery planning. As part of the Last Mile Science & Technology organization, you'll partner closely with other scientists and engineers in a collegial environment to develop enterprise ML solutions with a clear path to business impact. We are actively looking to hire scientists at various levels to innovate and lead on these problem areas. Successful candidates will have deep knowledge of competing machine learning methods for large scale predictive modelling and natural language processing, the ability to graduate models to production, the communication skills necessary to explain complex technical approaches to a variety of stakeholders and customers, and the ability to take iterative approaches to tackle big, long term problems.

Here is a glimpse of the problem spaces and technologies that we deal with on a regular basis:

• Organizing addresses into hierarchy in the presence of noisy, inconsistent, localized and multi-lingual user inputs. We do this at the scale of millions of customers for existing as well as emerging geographies, such as India, Spain, Australia, UAE. We make use of technologies like record matching, multi-modal architectures, named entity recognition, transformers and other language models for this problem
• Building a generic ML framework which leverages relationship between places to improve delivery experience by learning precise delivery locations and propagating attributes, such as business hours and safe places. This requires us to combine a variety of inputs (maps, delivery locations, defects) effectively, work in a multi-objective setting and exploit semantic as well as structural properties of places
• Explore semi-supervised learning, language modelling, data augmentation, active learning, information retrieval, ranking, etc. in the context of customer address text to build NLP models for address validation/suggestion, address parsing/spell-correction and geo-location learning.

• Lead a large scale high-impact project
• Review and provide feedback to junior scientists
• Publish work in internal and external top-tier conferences
• Provide inputs in team growth and goal planning
• Educate peers on science best practices
• Expand the use of ML into the orgaization.

• PhD or Masters in Computer Science, Mathematics, Statistics, or other quantitative field with exposure to statistical modelling and machine learning
• PhD with 3+ yrs or Masters with 8+ years of work experience in applied machine learning
• Ability to solve complex real-world problems in the industry
• Proficiency with at least one machine learning or statistical modelling library (Scikit-learn, R, etc.), and one programming language (Python, Java, C++, etc.)
• Excellent communication skills

• Publications in top ML conferences or journals
• Experience with a popular deep learning toolkit (TensorFlow, PyTorch, etc.)