At Inode Technologies, I wrote Python and SQL to validate, reconcile, and transform production datasets before analysis and reporting. I also created reusable checks for schemas, missing values, duplicates, and ranges.
I applied data-quality checks to machine-learning inputs and outputs, testing for drift, outliers, and inconsistent records. For time-series work, I engineered features and used CNN and LSTM models, evaluating performance on chronologically split data.
At Source One Management Services, on IBM projects, I tested FOSSE reservation, billing, and housekeeping workflows and verified fixes before release. I also used JMeter to assess reservation and booking transaction performance on a platform serving 500+ properties.
In a personal cloud data pipeline project, I built modular AWS Lambda stages orchestrated with AWS Step Functions, adding retry and error-handling logic. I also processed and reconciled datasets with PySpark on Databricks, bringing my data-validation experience into cloud data engineering.
