Software Development Engineer II @ Amazon
Working as backend developer (with occasional frontend tasks) in the Localization Core Team. Enabled localization of enriched product content.
In the industry since October 2016, before: M.Sc of Computer Science @ RWTH Aachen University
Experience in Machine Learning through various private projects and previous work experience as Student Research Assistant @ RWTH Aachen University
Joined a new team (formed in 2020) dealing with search-related issues in new and emerging marketplaces at Amazon. Dealing with indexing issues for new and emerging marketplaces at Amazon.
Delivered product pipeline to localize enriched product content. Scaled previous POC solution to 1M+ products per week. Integrated automation infrastructure to remove human effort from product selection. Introduced status reporting to track previous localizations and enable diving-deep in issues.
Working on automating the selection process for standard localizations. Integration with various content providers to gather evaluation data for products. Built a validation system to ensure product attributes fetched from an optimized data storage are in sync with the authoritative datasource.
Evaluation of speeding up Earth Mover's Distance calculations by using lower bound approximations as initial solution of the EMD. Analysis of linear program algorithms to solve the EMD, which are capable of using an optimal, but not feasible solution as initialization.
Implementation of a user interface and webservice operations to allow users of the Localization Project to
control the behavior of the tool for their input.
Implementation of a supporting application for the Localization Project, including implementing the backend and parts of the website for user input.
Implementing an RCNN to detect traffic signs in images. In a previous project I have used an off-the-shelf library to detect faces in images, this time I wanted to build the dectection engine on my own.