Rida Amjad

Senior Data Scientist

afiniti•  August 2020 - Present•  Islamabad

• As an AI Account Lead of LGU+ and SK Telecom (Afiniti’s South Korean telecom client), responsible for developing and implementation of machine learning models to predict a customer’s likelihood of upgrading a subscription. • Understanding the business problem, scope and risks and conducting an in-depth analysis such as correlations analysis with the target variable. • Meeting with clients on weekly or monthly basis to discuss business, scope and data related questions • Building of CRM using raw tables received from the client and extracting features from it to use in the model. • Designed feature engineering and feature selection automation tools to cater tables with high dimensionality through various techniques which turned out to be a huge time saver • Worked on a feature engineering function that automates creation of historic interaction features of a customer and previous change in subscriptions which added accuracy to the baseline model • Conduct a fortnightly code review for CRM codes written by the team of three data scientists for CRM, predictive modelling and deployment in production • Build a baseline model using different machine learning and deep learning model techniques (e.g. lightgbm, xgboost, PyTorch, etc.) and evaluate model validations. Discuss and implement ideas to improve this model to increase accuracies (or any appropriate validation tool to ensure model is predictive in test environment) • Building and implementing a runtime modelling strategy that ensures the efficient pairing of agents with customers to maximize KPI’s such as increase in sales, client retention, etc. • Provide analytic insights and advice to maximize the target and in some cases share these with the client • Use state of the art tools such as deep checks, great expectations, evidently, and design custom-built tools for daily checks on data’s integrity, data drift and model performance • Served as an Applied AI Trainer for this quarter and trained AI members on automation of feature engineering, runtime data and performance checks and exploratory analysis required for a new deployment • Proficient in Python (i.e. Pandas, Numpy, scikit-learn, etc), R (dplyr, ggplot, lubridate, knitr, etc.), Postgresql, Power BI, Azure, and MySQL.

University of Dundee

Business Administration, MS•  September 2015 - September 2016•  null: 5

Got a Distinction.