Manobhiram  ·  Data Analyst  ·  5+ yrs

Overview: This project involved developing analytics solutions, reports, and dashboards using Microsoft Power BI Desktop and Azure services for a major retail client. Responsibilities: Developed Power BI graphical and visualization solutions, incorporating DAX functions for complex calculations and measures. Utilized SSIS to create ETL packages for validating, extracting, transforming, and loading data into data warehouses and data marts. Implemented Azure Data Factory v2 for data ingestion and control into Azure data lake store. Designed and deployed SSAS multidimensional cubes for analytic research and report generation. Increased query performance by monitoring, tuning, and optimizing indexes using SQL Profiler.

Overview: This project focused on developing and deploying Power BI reports and dashboards, and implementing SSIS packages for a retail client. Responsibilities: Published Power BI reports on dashboards to Power BI services and pinned visuals for enhanced visualizations. Utilized Power Query to pivot and unpivot tables for data cleansing and manipulation. Created SSIS packages to load various Dimensions and Fact tables based on complex business requirements. Established DAX while creating data models in Power BI and converted SQL Query code into DAX for reports. Created conceptual, logical and physical data models using best practices.

Overview: Developed a Python-based command-line utility to save and restore Linux desktop settings, enabling seamless configuration transfer. Responsibilities: Built a utility that allows users to export and import desktop settings between computers. Implemented support for switching between multiple saved profiles, offering flexibility for users.

Overview: Developed and orchestrated a complete ML pipeline using MLflow to streamline MLOps for efficient model lifecycle management. Responsibilities: Automated data preprocessing, model training, evaluation, and deployment, ensuring reproducibility and version control. Integrated MLflow for tracking experiments, managing model versions, and monitoring models in production. Enabled CI/CD for machine learning models, improving collaboration between data science and DevOps teams. Optimized deployment and resource utilization, reducing the time to production for machine learning workflows.