• Developed an image retrieval system using autoencoders, achieving above 90% retrieval accuracy for similar images across a dataset of more than 50,000 samples.• Improved machine learning model’s R² scores by 100% for over 100 different datasets by implementingautomated code using Python and libraries like Scikit-learn, Pandas, and NumPy.• Implemented advanced techniques like boosting for model building and dimensionality reduction methods such as PCA and Auto-Encoders, enhancing productivity by 40%.• Leveraged SQL in GCP BigQuery for efficient data extraction and processing, achieving a 30% reduction inquery times.

Responsibilities : - • Utilized Python, PySpark, and HDFS to create expandable data workflows for seamless data management improving data processing efficiency by 20-30%• Sustained and refined data storage on AWS S3 guarantees data accessibility also reducing overall costs in the process• Created containerized workflows with docker for maximum portability and reliability.• Created efficient SQL scripts for data extraction and processing from Oracle databases.

Responsibilities : - • Streamlined data processing done through strategically designed Azure Data Factory pipelines reducing processing time by 10-20%.• Applied Python for data manipulation and PowerShell for streamlining.• Ensured proper data management in HDFS, Azure SQL, and SQL Server allows for efficient storage and retrieval.• Deployed real-time data processing with Azure Functions as part of projects and multiple Proof of Concepts.