- Develop a Python program for downloading, processing, and visualizing topographic and environmental data. This project is part of deploying Clean Nature's AI-based tool, guIA, to optimize road de-icing practices for municipalities.- Develop a Python-based method for regionalizing flood quantile modeling using GEV density networks. This project involves integrating GEV distributions into an existing Python program, calibrating and validating the algorithm, and generating flood risk maps. The work supports advanced hydrological modeling to improve flood prediction and risk assessment.

- As a Data Scientist at DMVET Veterinary Hospital, I used Python with deep learning and transformer models to forecast daily emergency visits, enhancing resource management and efficiency. My role involved fine-tuning and deploying these models to improve staffing and resource allocation decisions.- Utilized unsupervised ML for record linkage, integrating data sources for comprehensive patient records.

- Database Management and Maintenance: Proficiently manage and maintain databases, specializing in CKAN, ensuring data accuracy and accessibility.- Dashboard Development: Create dynamic dashboards to visualize data trends and insights, leveraging CKAN to enhance information dissemination.- Artificial Intelligence in Health: Contribute expertise to cutting-edge research in the field of artificial intelligence and digital health.- Statistical Learning and Modeling: Lead efforts in developing advanced statistical learning methods, artificial intelligence algorithms, and mathematical/statistical models. Focus on applications in infectious disease analysis and climate change risk assessment.

- Developed a CRM Sentiment Analysis Platform with Django and Flask, integrating NLTK and Stanford NLP to analyze English and non-English social media content, improving content strategies from user feedback.- Developed a dynamic fleet tracking portal with driver incentive analysis capabilities using Python's PySpark and Power BI, enabling the operations department to efficiently analyze incentive plans and track fleet performance.- Implemented a machine learning-based system to detect dual drivers collaborating with competitors, utilizing deep learning models to analyze performance driver attributes. Leveraging incentive plans to encourage their alignment with our operations led to a 10% increase in market share within three months.- Implemented probabilistic modeling using machine learning techniques to predict churn drivers and devised compelling incentives to mitigate churn, resulting in a 25% reduction in churn rate in a year.- Developed a real-time dashboard in Power BI using Python scripts to track fleet health every 15 seconds, enabling dynamic demand forecasting and enhancing operational efficiency.- Implemented EM algorithm for driver segmentation, enabling customized management of incentives and plans based on behavior and improving operational efficiency.- Developed an interactive Power BI dashboard for fleet health monitoring, enhancing driver quality control and operational oversight.- Implemented SEO optimization for the Passenger App using "SEOGenius" or "NLP-SEO Analyzer," enhancing search engine visibility and user engagement.- Implemented a machine learning-based approach in Python for Fraud Detection, combining anomaly detection in transaction data with behavioral analysis of users, enhancing fraud detection capabilities.

Teaching assistance of Machine Learning and Business Analytics master courses. I was responsible for:- Set up visual aids, equipment, and classroom displays to support teacher's lesson delivery.- Assessed student assignments to check quality and completeness before submission for grading.

- Utilized NER to capture prescription entities, matched with doctor information, and applied association rules with the A-Priori Algorithm in R for health insurance fraud detection in prescriptions.- Implemented an LSTM predictive model in Python to estimate the education duration of master students at SBU, achieving an RMSE of 2.24 months, aiding in identifying students with extended graduation periods in the context of free education in Iran.- Managed the development and execution of an employee satisfaction survey from inception to completion, providing a comprehensive analysis for Sina Darou Labs. (One of the pioneering Sterile Eye and Respiratory Aerosol Manufacturers in Iran).

- Development of the Iranian households TV Audience Measurement survey using a people meter device (TAM), implementing complex sampling methods to prevent household churns, and creating automated reports and dashboards in Power BI.- Implemented computer vision and NLP techniques for monitoring advertisements, extracting time, content, caption, etc., to categorize and sell information to consulting companies.- Orchestrated the development of a television advertising pricing platform to optimize advertising strategies based on TAM data and pricing information.- Provided statistical consulting for the retail census, leveraging extensive consumer surveys to deliver insights on demographics, attitudes, media, and purchasing.