- Modeling job rotation and estimating the effectiveness of proposed employees’ arrangement using Markov chainanalysis.- Analyzing employees’ job rotation in organizational positions in different levels. - Evaluation of employees in different organizational positions for a 5-year period. - Checking through the whole database of employees in SAP including performance, personal information, recruitmentsanddischarges, and overall functions, and creating related infocubes. Connecting SAP database to Power BI todevelopmanagerial dashboards for the employees. The dashboards were all created in Power BI framework and includeddescriptive insights using abovementioned HR indicators. Data cleansing and modeling were also undertakenusingPowerBI. - Designing analytics dashboards for decision making process in HR management using indicators such as jobsatisfactionand its influence on the overall performance of employees. These dashboards led to a better management of servicedistribution among the employees, reducing resource misspend and increasing job satisfaction.- Creating a code in Python for departure risk assessment which finds employees’ reasons for departure usingaclusteranalysis and predicts every person’s departure probability for the next work year with an 85-percent precision. Thisassessment was based on a classification of departure reasons from a database obtained fromseveral years of annual reports into two main reasons namely migration or better job opportunities. - Developing a code in Python using time series data reframed in Markov chain to estimate personnel’s next stepsandpossible progress in skills and organizational hierarchy. It also yielded good insights on role reversals andpossiblereplacements. This tool ended up to be greatly beneficial since it helped the administration to learn about anypossibledepartures in advance, which had always imposed unforeseen costs to recruit and train new, usually underpreparedemployees.

During my tenure as a freelance collaborator with Opark, I led a strategic project aimed at enhancing customer segmentation through RFM (Recency, Frequency, Monetary) analysis. By meticulously analyzing transactional data and leveraging advanced algorithms such as K-means clustering, we segmented customers into distinct groups, enabling Opark to identify and prioritize their most valuable customers. This segmentation facilitated targeted marketing initiatives and personalized offers, ultimately driving increased customer satisfaction, loyalty, and enabling more efficient resource allocation. Through the seamless integration of Python libraries for data processing and analysis, we delivered actionable insights that empowered Opark to optimize its marketing strategies and enhance its competitive edge in the online parking service industry. This project underscored the transformative impact of data-driven strategies in driving tangible business outcomes and fostering sustainable growth within Opark's dynamic operational landscape.

Working as a fraud detection analyst finding misleading submitted information about products uploaded in Divar online Ecommerce platform, designing a model to gather and regroup data from scams to find common factors and patterns among these fake accounts, the results of which helped system to change its registration and identity verification exercises.• Development and enhancement managerial dashboards finding the most important KPIs influencing the sales, to establish a more effective investment in order to achieve a higher profitability. The managerial dashboards for sales data helped the administration find out investments in which sales sector in each province of Iran had a better revenue for the company.

Working as a digitalization specialist, designing online management dashboards. Providing quantitative and qualitative analyses and submitting analytical reports to improve the processes, using Python, machine learning, decision tree, clustering, random forests, classification & regression, etc.

• Assigned with data cleansing and creating require tables for SQL, dealing primarily with sales data derived from each branch store to raise sales efficiency.• Designing a sales model using association rules algorithm in R, clustering, and machine learning models which revealed a significant relation between the placement of products and sales performance. The project resulted in predicting how to place the products to achieve multiplied sales.• Designing online sales reports in Power BI using various KPIs. The procedure began with descriptive dashboards ending with predictive analytical dashboards using predict methods in Power BI and Dex.