Meisam G.
Data Scientist
Experience
Jun 2023 - Present
2 years 7 monthsData Scientist
Geeks Ltd
- 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 monthsData 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 yearData 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
NativeGerman
AdvancedEnglish
AdvancedEducation
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
Need a freelancer? Find your match in seconds.
Try FRATCH GPT More actions
Similar Freelancers
Discover other experts with similar qualifications and experience
- English
- Deutsch
2025 © FRATCH.IO GmbH. All rights reserved.