Greg Mehdiyev's Projects
Playing around with Linux Kernel
The goal of this project is to design and train deep convolutional neural networks using PyTorch. I will design a deep net architecture to classify (small) images into 100 categories and evaluate the performance of the architecture by uploading the predictions to the Kaggle competition.
GrocerySegmentAI is a platform as a service for grocery businesses. This product provides a way for these businesses to upload customer data and receive accurate predictions on which customers will buy a specific product on their next order.
Open-source music streaming platform that uses microservices, Docker, Flask and DynomoDB.
The goal of this project is to build a instance segmentation model that is able to detect and segment airplanes from the satelite images from the airporst around the world.
A cloud service that collects buyer property preference data in real-time, puts it through a data streaming pipeline and feeds it into the compute layer, where data is analyzed in the ML pipeline and property recommendations are adjusted based on new buyer preferences.
The task for this project is to segment a sequence of English characters into the most likely word sequence.