Utilized a Large Language Model (LLM) at WordUp, tailored to enhance vocabulary learning by understanding and generating contextual examples, improving personalized learning experiences
Utilized machine learning to analyze user behavior and predict churn, identifying key engagement trends that led to a 15% increase in user retention and satisfaction
Designed and analyzed A/B experiments for new app features, measuring engagement and retention impact; provided actionable insights that improved user activation by 15%
Developed a Customer Lifetime Value (CLTV) prediction model, leading to a 10% increase in average CLTV through targeted retention efforts
Developed a Bayesian Marketing Mix Model to optimize budget allocation across digital and offline channels, improving ROI by 12%
Designed and implemented a high-performance Python ETL pipeline, optimizing CPU and I/O utilization and streamlining data cleansing logic, resulting in a 30% reduction in processing time
Reduced abandonment rate by 10% by analyzing user onboarding behavior and identifying key drop-off points
Feb 2020 - May 2023
3 years 4 months
Data Scientist
Mellat Bank
Developed a news trading algorithm using Fin-BERT and GPT3 and sentiment analysis to predict short-term price movements in Forex and Crypto markets, achieving a 20% improvement in trading signal accuracy
Built an ensemble model combining XGBoost with a deep auto encoder to detect anomalies in credit card transactions in real time, reduced false positives by 25% and improved fraud recall by 15%
Implemented a customer churn prediction model, reducing churn by 15% and enhancing customer retention strategies
Built recommendation engines to suggest banking products to customers, enhancing cross-selling opportunities by 22%
Developed a sentiment analysis model using BERT to analyze Instagram comments, achieving over 85% accuracy and providing actionable insights for customer engagement strategies
Engineered a forecasting model using Deep AR to predict credit usage with an R-square value exceeding 80%, optimizing financial planning
Designed and implemented a credit risk scoring system that improved early identification of high-risk applicants and reduced default rates by 12%
Mar 2019 - Feb 2020
1 year
Data Scientist
Hosh and Dansh
Created machine learning models to predict inventory needs, reducing holding costs by 15%
Built models to predict customer churn, enabling proactive retention efforts that decreased churn rate by 12%
Implemented algorithms that adjusted pricing based on demand forecasting, increasing revenue by 10%
Languages
Persian
Native
German
Advanced
English
Advanced
Education
Science and Culture University, Iran
Master of Financial Engineering and Risk Management · Financial Engineering and Risk Management · Tehran, Iran, Islamic Republic of
IU International University of Applied Sciences, Berlin
Master of Data Science · Data Science · Berlin, Germany
Azad University, Iran
Bachelor of Industrial Engineering · Industrial Engineering · Tehran, Iran, Islamic Republic of