Rutvi Macwan Email & Phone Number

Melbourne, Victoria, Australia

• Personalised Modelling: Currently developing a Personalised Recommendation Model to allocate tailored offers to customers, closely collaborating with the commercial teams. • Predictive Modelling: Achieved generosity savings of $18M through the implementation of Predictive Machine Learning Modelling (Random Forest) by identifying risky customer behaviour on average 3 months before the manual intervention.• Data Preprocessing & Feature Engineering: Improved model accuracy from initial 59% to 88% by employing advanced data preprocessing methodologies, including one-hot encoding, label-encoding, strategic class imbalance techniques, and different feature engineering technologies.• Data Analysis: Boosted utilisation rate of a recommendation model from 6% to 15% by proactively identifying improvement opportunities through data analysis using pySpark and SQL.• Cross Functional Team Collaboration: Collaborating with diverse business stakeholders including Data Engineers, MLOps, RPA, and other commercial stakeholders to streamline end-to-end model delivery.• Data Migration & Leadership: Led a team of 20 people for the successful data and model migration to a new DataBricks platform under tight deadlines.• Technical Documentation: Documented Technical details of Model Development journey along with reasoning for method selections, feature engineering and future improvement opportunities. Documented Databricks best practices, including cluster configuration, to support the team for optimised DataBricks utilisation.• Presentation: Presented key ideas of machine learning modelling to various non-technical stakeholders including the CEO of the company.

Melbourne, Victoria, Australia

• Overview: ​​Developed advanced AI functionalities within a product, leveraging cutting-edge technologies including Text Summarization, Amazon Textract and OCR to enhance the product features. Performed feature testing using Docker and Wit.AI; and model deployment using AWS endpoint. • AutoML: Developed various Automated Machine Learning Modelling technologies using AutoML, AutoARIMA and automated clustering algorithms to add functionalities in a product.

Victoria, Australia

• Conducted in-depth analysis and exploration using Python on 8M rows of data gathered from IoT sensors. Achieved an accuracy of 90% by developing an unsupervised AutoEncoder model for anomaly detection in cars.

Bangalore

Bangalore

Mysore, Karnataka, India

Master Of Science - Ms, Data Science, 3.7

Bachelor Of Engineering (B.E.), Computer Engg., 8.55