Engineered a predictive analytics model using Random forests to anticipate and mitigate delivery bottlenecks, achieving a 95% accuracy rate and enhancing overall delivery efficiency by 20%
Launched a real-time anomaly detection system employing autoencoders, which allowed the rapid identification and resolution of discrepancies in the logistics chain, reducing overall operational hiccups
Produced daily and weekly commercial reports using R, assisting in data-driven decision making
Designed and deployed an RShiny application tailored for breeding data visualization and analysis
Software ML Engineer
Amazon
06.2022 - 04.2023
Developed and deployed scalable ML models to optimize last mile delivery routes, reducing delivery time by 16%
Collaborated with cross-functional teams to integrate predictive models into the existing infrastructure
Led end-to-end implementation of a real-time anomaly detection system, promptly identifying and addressing delivery disruptions
Created a robust testing and validation framework ensuring model accuracy and integrity post-deployment
Implemented an A/B testing framework to assess the effectiveness and performance improvements brought about by new model iterations
Utilized Random Forests and Gradient Boosting Machines to analyze and score driver behaviors, enhancing safety and efficiency
Data Scientist
Teave Tech LLC
09.2021 - 12.2021
Conceptualized in Data transport, storage and cleansing techniques, microservice development, implementing Machine learning models and supporting deployment of a production SaaS platform
Built a simulation model to improve demand planning for store replenishment
Found 32% of error in the forecast by using XGBoost vs Rolling mean (Python stack)
Set up Power BI dashboards to identify product features that correlated with customer attrition, leading to proactive interventions that reduced monthly churn by 4%
Built data visualizations using SQL and Tableau for product KPIs that reduced manual reporting work by 8 hours weekly
Developed ETL pipelines using Pyspark for processing large datasets on a cluster
Data Scientist
SGS Tekniks
11.2014 - 05.2018
Analyzed sales figures, market research, logistics or transport data
Converted data into actionable insights by predicting and modeling future outcomes
Coming up with solutions to costly business problems
Built prediction models for Probability of Purchase, predictive Order Date, Quantity of Purchase for SGS customers improving model's AUC by 10%, RMSE by 12% and RMSE by 5% respectively
Consolidated vendors and created a volume advantage, resulted in a 12% overall reduction in marketing costs
Utilized TensorFlow's high-level APIs, such as Keras, to streamline the model development process and improve productivity
Conducted data preprocessing and feature engineering using TensorFlow's data manipulation and transformation functionalities
Implemented a long-term pricing experiment that improved customer lifetime value by 32%
Education
Master of Science - Data Science & Management
The State University of New York
Bachelor of Technology - Electrical and Electronics Engineering
SRM University
Skills
Technical: PowerBI, Spotfire, Tableau, JIRA, Agile development, GitHub, Leadership experience, Software Development Lifecycle, HEOR, Rest API, Beautiful Soup, BigQuery, PostgreSQL, Domino Data Lab, GIT, GITLAB, AWS, Microsoft Excel, Matlab
Machine Learning: Regression, Classification, SVM, Decision tree, Random Forest, Naïve Bayes, Gradient Boosting, Computer Vision, Neural Network, Time Series, Statistics, Bayesian Statistics, Probabilistic modeling, Nonlinear Dynamics, Hierarchical models, A/B Testing, Python, R, Data Science, Statistics, SQL, ETL Pipeline, , Regression, Applied Mathematics, CI/CD, Probability, PySpark , Self-taught C studying in-depth intricacies on data structures