Chaitrali Pardeshi

Supported LOI BL endorsement processes and DPID creation activities for logistics and shipping operations.
Contributed to vessel performance tracking and analytics, leveraging SAP ERP for real-time operations monitoring and report automation.
Generated and maintained Excel-based reports to enhance operational transparency, data-driven decision-making, and compliance requirements.
Uploaded Statements of Facts (SOF) for vessels on the MILA portal of BPCL, ensuring timely and accurate logistics documentation.
Collaborated cross-functionally with technical teams for documentation, process streamlining, and audit readiness in line with industry best practices.
Gained hands-on exposure to SAP ERP modules, operational reporting, and logistics workflows, strengthening technical and analytical skills in a core engineering domain.

- Served as Engineering Apprentice at Bharat Petroleum Corporation Limited from Dec 2024 to Mar 2025, supporting LOI BL endorsement processes and DPID creation activities for logistics and shipping operations.
- Contributed to vessel performance tracking and analytics, leveraging SAP ERP for real-time operations monitoring and report automation.
- Generated and maintained Excel-based reports to enhance operational transparency, data-driven decision-making, and compliance requirements.
- Uploaded Statements of Facts (SOF) for vessels on the MILA portal of BPCL, ensuring timely and accurate logistics documentation.
- Collaborated cross-functionally with technical teams for documentation, process streamlining, and audit readiness in line with industry best practices.
- Gained hands-on exposure to SAP ERP modules, operational reporting, and logistics workflows, strengthening technical and analytical skills in a core engineering domain.

Developed and deployed predictive models (Logistic Regression, Decision Trees) for customer churn and risk prediction in BFSI portfolios, improving retention by 18%.
Built and maintained Power BI dashboards for credit risk, loan portfolio health, delinquency trends, and default probabilities, enabling timely and data-driven decisions.
Designed fraud detection frameworks for payment transactions and insurance claims using anomaly detection, reducing fraudulent activity by 12%.
Partnered with underwriting and claims teams to create risk scoring models for insurance applications, enhancing assessment accuracy by 15% and accelerating claims processing.
Implemented Customer Lifetime Value (CLV) models to drive cross-sell/upsell opportunities and improve retention in financial and insurance products.
Automated ETL pipelines using Python and SQL, increasing data processing speed by 40% and ensuring accuracy in financial reporting.
Applied K-Means clustering for customer segmentation in insurance marketing, enabling targeted campaigns and improving engagement rates.
Conducted A/B testing and statistical analysis to optimize BFSI product features, leading to a 12% boost in adoption.
Performed data wrangling, feature engineering, and model optimization (cross-validation, hyperparameter tuning) to improve accuracy and reduce false positives.
Delivered actionable insights through data storytelling and executive-level dashboards, supporting strategic decision-making for BFSI stakeholders.

- Developed and deployed predictive models (Logistic Regression, Decision Trees) for customer churn and risk prediction in BFSI portfolios, improving retention by 18%.
- Built and maintained Power BI dashboards for credit risk, loan portfolio health, delinquency trends, and default probabilities, enabling timely and data-driven decisions.
- Designed fraud detection frameworks for payment transactions and insurance claims using anomaly detection, reducing fraudulent activity by 12%.
- Partnered with underwriting and claims teams to create risk scoring models for insurance applications, enhancing assessment accuracy by 15% and accelerating claims processing.
- Implemented Customer Lifetime Value (CLV) models to drive cross-sell/upsell opportunities and improve retention in financial and insurance products.
- Automated ETL pipelines using Python and SQL, increasing data processing speed by 40% and ensuring accuracy in financial reporting.
- Applied K-Means clustering for customer segmentation in insurance marketing, enabling targeted campaigns and improving engagement rates.
- Conducted A/B testing and statistical analysis to optimize BFSI product features, leading to a 12% boost in adoption.
- Performed data wrangling, feature engineering, and model optimization (cross-validation, hyperparameter tuning) to improve accuracy and reduce false positives.
- Delivered actionable insights through data storytelling and executive-level dashboards, supporting strategic decision-making for BFSI stakeholders.

Developed and validated machine learning models for real-world datasets, including Google stock price prediction and Titanic survival analysis, achieving over 90% precision in both cases.
Applied advanced feature engineering techniques to extract relevant attributes from raw data, leading to an 18% improvement in model accuracy.
Conducted correlation analysis and handled missing values using imputation strategies.
Built end-to-end machine learning pipelines using Python, automating data preprocessing, model training, and evaluation.
Performed hyperparameter tuning and cross-validation to optimize model performance and reduce overfitting.
Communicated results through clear visualizations using Matplotlib and Seaborn, improving stakeholder understanding of model behaviour and key drivers.
Delivered two complete data science projects with documented code, Jupyter Notebooks, and summary reports, showcasing strong project ownership and technical communication.

- Developed and validated machine learning models for real-world datasets, including Google stock price prediction and Titanic survival analysis, achieving over 90% precision in both cases. Utilized algorithms such as Linear Regression and Logistic Regression with Scikit-learn.
- Applied advanced feature engineering techniques to extract relevant attributes from raw data, leading to an 18% improvement in model accuracy. Conducted correlation analysis and handled missing values using imputation strategies.
- Built end-to-end machine learning pipelines using Python, automating data preprocessing, model training, and evaluation. Streamlined workflows enhanced model usability and scalability for future use cases.
- Performed hyperparameter tuning and cross-validation to optimize model performance and reduce overfitting.
- Communicated results through clear visualizations using Matplotlib and Seaborn, improving stakeholder understanding of model behaviour and key drivers.
- Delivered two complete data science projects with documented code, Jupyter Notebooks, and summary reports, showcasing strong project ownership and technical communication.