Data Engineering Developer in Toronto, ON, Canada

- The aim of the project was to improve the click-through rates of mobile push notifications
- The introduction of the algorithm resulted in 15% improvement in the Mobile Push Notification Click-Through-Rate
- I decided to achieve this by developing a machine learning algorithm that predicted the optimum contact time for every user
- The algorithm was developed in Python and was trained using scikit-learn
- Obtaining training data required the use of HIVE and PySPARK
- I have successfully implemented the algorithm into the marketing production environment and instructed marketing analysts on how to use it

- The aim of the project was to increase the click-through rate of eBay coupon campaigns via the use of machine learning
- The development of the algorithm was successful and it was measured to generate 20% lift in click-through-rate and iGMB
- The early version of the algorithm was based on the multi-armed bandit. Later versions made use of contextual NLP-based multi-armed bandit
- The algorithm was developed using a combination of Teradata SQL and Python
- I have also developed an interactive Tableau dashboard in order to monitor the function of the algorithm and to measure the KPI lift that the algorithm was bringing

- The aim of the project was to develop a machine vision algorithm capable of detecting pedestrians in front of a vehicle by analyzing the input from the vehicle camera
- The algorithm is now fully functional and is embedded into a number of newer vehicle models including the GM 2019 Chevy
- The machine learning algorithm we decided to use was ada-boost cascade classifier combined with deep neural network
- We wrote the training application from scratch in C++. Training had to be multithreaded in order to be efficient.
- Testing and validation was done in Python
- A large database of annotated video data was used for algorithm training

- Precise prediction of the total final loss after the default of a client is a key to reducing risk associated with different loan products
- I have developed a model that relied on the loan to value ratio and the value of the collateral
- Development was done using a combination of SAS and MS SQL
- Development of the model required extensive cleaning of the data and data quality testing

- Involved development of a recommendation engine based on collaborative filtering model.
- The engine was capable of recommending even the products that a given customer did not necessarily buy in the past
- The solution was implemented in PySPARK and was based on SPARK ML

- Developed an ETL in PySPARK for transfer of data from Redshift into an S3 data lake
- Developed code for customer-level data aggregation and historization
- Assessed data quality, investigated and remediated data quality issues