Vatsal Patel Email & Phone Number

Little Rock, Ar, Us

Little Rock, Arkansas, United States

Quincy, Massachusetts, United States

• Engineered high-impact business solutions for the E-Commerce and Digital Team, developing scalable machine learning models, utilized 25TB of historical sales and search data, applying GenAI (Large Language Models) yielding powerful product and search recommendations• Developed Transformer based ML models for the Marketing Team, utilizing PyTorch resulting in contributing to $1M incremental revenue within the first year, reducing operational costs by 15% through an automated system to decide which ads to show based on user profiles• Designed and deployed a session-based recommender system leveraging transformer models and GPU acceleration via Nvidia RAPIDS. Implemented distributed computing with Dask, slashing data preprocessing times by 60% and boosting recommendation relevance by 35%• Crafted an advanced payment fraud detection system utilizing LSTM networks for temporal feature extraction, processing 1M transactions weekly, resulting in a 15% reduction in fraudulent transactions, saving the company an estimated $5M annually

Ahmedabad, Gujarat, India

• Innovated a hybrid machine learning algorithm by fusing Decision Tree and Linear Regression techniques, creating a novel predictive model that dynamically segments data and optimizes feature selection. This advanced approach reduced Mean Squared Error by 12% on average, outperforming standard models by 20% in accuracy and 15% in processing speed across diverse datasets• Developed an improved version of the Decision Tree regressor by combining the Linear Regressor, for finding optimal predictors by repeatedly breaking the data into 2 segments and fitting a linear regressor thereby reducing MSE. (Python Modules: pandas, numpy, scikit-learn, matplotlib).• Performed experiments on various standard datasets to compare the performance of Linear Decision Tree Regressor model with other standard regressors and found the model to outperform other models in terms of accuracy and overall predictions.• Displayed understanding of limitations linear regression models and decision tree models in solving regression problems.• Link to the paper: https://www.researchgate.net/publication/353546682_Linear_Decision_Tree_Regressor_Decision_Tree_Regressor_Combined_with_Linear_Regressor

Ahmedabad, Gujarat, India

• Architected and implemented a high-performance data integration pipeline using Alteryx, unifying 15+ data sources into a AWS S3 using AWS Data Lake Formations improving data retreival times by 40%, reducing data preparation times by 70% supporting data-driven decision making• Facilitated data-driven insights for Merchandising Team, implementing cloud-native data solution, leveraging AWS and Databricks, collaborating with cross-functional teams, monitoring KPI’s using dynamic dashboards, reducing annual infrastructure costs by $25,000• Implemented data quality checks for the Finance and Accounting Team to ensure data integrity, consistency and accuracy of data, using Python and AWS EMR tools such as EMRFS and S3DistCp, implementing tests on the source data to ensure data quality• Led research, strategic planning, and experimentation to design infrastructure, utilizing AWS and services lie AWS Lake Formation, Dynamo DB, Elastic Cache, AWS Redshift, and AWS Step Functions, optimizing performance and improving response time, reducing latency by 20%

Ahmedabad, Gujarat, India

• Engineered an inventory forecasting model using LSTM neural networks in TensorFlow, driving a 6:1 inventory turnover ratio for retail clients. Led cross-functional collaboration with key stakeholders to optimize supply chain efficiency and significantly reduce carrying costs• Delivered Exploratory Data Analysis across 5 product lines, leveraging Pandas, Seaborn to analyze 50M+ transactions, architected anomaly detection model on AWS SageMaker using Random Cut Forest algorithm reducing fraud by 28%, directly impacting bottom line

Master Of Science - Ms, Artificial Intelligence, 3.8

Master'S Degree, Artificial Intelligence

Bachelor'S Degree, Computer Science

Education record

Studied at Delhi Public School from Middle School to High School.