Developer Experience and Development Environment

Ads Optimization and Modeling

PlaceSnap team:

• developing one of the core location intelligence products: machine learning powered, multisensor Snap-to-Place system

• working on improving location prediction accuracy

• extracting insights from data obtained from our partners

• Implemented CopyCatcher (https://github.com/nataliest/copycatch), a computer vision powered service that quickly filters almost 600GB of Open Images Dataset files to find a copy of an uploaded image

• Set up a Spark cluster for running image upload and comparison scripts, AWS Elasticache/Redis for storing image tags and metadata, and S3 for storing raw image files; utilized Bash and Python libraries

• Compared performance of databases (Redis vs. Neo4j) to find the best fit for cataloguing the dataset

• Created a web application showing detailed image upload and comparison statistics using Flask and HTML/CSS.

Object-Oriented Programming, Algorithms, and Data Structures in Python and C++;

Fundamentals of Electric Circuits: DC and AC circuit analysis.