• Led the development of a Python-based anomaly detection tool, automating abnormality location detection in audio clips, eliminating the need for manual intervention and reducing processing time by 90%. • Achieved high accuracy in anomaly detection tool using bandwidth filters, improving monitoring of over 10,000 hours of poultry audio data. • Enhanced the anomaly detection tool by integrating support for user-defined frequencies with adjustable thresholds, enabling highly precise anomaly detection across hundreds of audio patterns. • 80% faster analysis achieved by automating anomaly score calculation, eliminating the need for manual grading thousands of segments. • Enhanced an anomaly detection tool by implementing automated preprocessing, delivering thousands of correctly formatted datasets to streamline machine learning workflows

• Collaborated with a six-member Scrum team to deploy ”LabelMe,” an AWS Amplify-powered platform built with Python to convert audio files into spectrograms for advanced anomaly detection. • Designed SQL workflows to process and export anomaly labels for millions of audio records, ensuring accurate data updates across 25 database tables with seamless AWS Lambda integration. • Developed SQL workflows, leveraging AWS Lambda to import S3 audio file paths and anomaly labels from 25 database tables, ensuring efficient retrieval of audio files from an S3 bucket. • Authored clear and precise code documentation, streamlining onboarding and ensuring effective communication for future developers