Data Scientist Graduate Intern
San Ramon, California, Us
• Built in-house machine learning predictive analysis tool leveraging proprietary SequentialSimilarity Modeling (SSM) technology that indicates remaining functional life of a failing asset,resulting in 280K savings in yearly maintenance costs per plant• Developed data pipeline to automatically process raw data, feed it into SSM model, andgenerate the best model parameters, reducing manual work by 92% from 14 days to 1 day• Proved SSM capability with proof-of-concept analytic, achieving 70% improvement overstatistical median, gaining management's approval for continued work on the project