Shripal Shah Email & Phone Number

Bengaluru, Ka, In

Developed and deployed a predictive churn model with 92% accuracy and 85% precision using Python,TensorFlow, and Scikit-learn, enabling a 25% reduction in customer attrition.• Designed a distributed data pipeline leveraging Apache Kafka and Hadoop to process 10+ TB of customerinteraction data daily, reducing latency by 40% for real-time insights.• Built and optimized ETL workflows using Apache Airflow to ingest, transform, and load structured andunstructured data into… Show more Developed and deployed a predictive churn model with 92% accuracy and 85% precision using Python,TensorFlow, and Scikit-learn, enabling a 25% reduction in customer attrition.• Designed a distributed data pipeline leveraging Apache Kafka and Hadoop to process 10+ TB of customerinteraction data daily, reducing latency by 40% for real-time insights.• Built and optimized ETL workflows using Apache Airflow to ingest, transform, and load structured andunstructured data into Snowflake, improving data availability by 30%.• Conducted advanced feature engineering and exploratory data analysis (EDA) using Pandas, NumPy, andMatplotlib to identify 15+ key factors influencing churn.• Deployed scalable machine learning models on AWS SageMaker, integrating with S3 and EC2 for efficienttraining and inference, reducing infrastructure costs by 20%.• Integrated a large language model (LLM) pipeline using Hugging Face Transformers to analyze customersentiment from support logs, enhancing feature generation for predictive models and improving 5% accuracy.• Implemented real-time RESTful API endpoints for predictive insights, ensuring latency under 200ms andseamless integration with customer engagement systems. Show less

San Francisco Bay Area

• Designed and implemented a comprehensive data engineering and analytics pipeline to power an AI grantwriting tool, enabling streamlined funding applications and delivering measurable efficiency gains.• Engineered and deployed an AI-powered grant-writing Minimum Viable Product (MVP) using Large LanguageModels (LLMs), achieving an 80% reduction in proposal completion time and simplifying the funding applicationprocess.• Architected and implemented a Retrieval-Augmented… Show more • Designed and implemented a comprehensive data engineering and analytics pipeline to power an AI grantwriting tool, enabling streamlined funding applications and delivering measurable efficiency gains.• Engineered and deployed an AI-powered grant-writing Minimum Viable Product (MVP) using Large LanguageModels (LLMs), achieving an 80% reduction in proposal completion time and simplifying the funding applicationprocess.• Architected and implemented a Retrieval-Augmented Generation (RAG) pipeline leveraging LangChain andFAISS, integrated with Google Vertex AI to index 100,000+ grant documents, reducing manual research time by90% while enhancing response accuracy.• Developed a scalable ETL pipeline using Apache Spark to process and transform 20+ GB of unstructured grantdata from diverse sources, reducing data preparation time by 40% and ensuring seamless integration with theRAG pipeline.• Automated embeddings generation for grant documents via Python-based Cloud Run functions on Google CloudPlatform (GCP), utilizing Google Cloud Storage (GCS) for embedding storage and improving pipeline retrievalspeed by 30%.• Built and optimized RESTful APIs with Flask to connect AI models and Firebase databases, cutting response timesby 30% while deploying scalable backend services on AWS Elastic Beanstalk, EC2, and S3 to enhance systemperformance by 25%.• Configured Cloud Firestore database for real-time storage of user queries and grant-related answers, improvingdata retrieval and response accuracy by 60% through advanced analytics and dynamic updates. Show less

Karnataka, India

• Designed and optimized Python scripts to automate data ingestion, transformation, and analysis. Managed ETLworkflows using Apache Airflow, achieving a 30% reduction in processing time and increasing data extractionrates by 25%, significantly improving pipeline efficiency.• Leveraged advanced SQL queries to extract and analyze large datasets from relational databases, providingactionable insights that drove a 10% increase in user engagement metrics for the platform.•… Show more • Designed and optimized Python scripts to automate data ingestion, transformation, and analysis. Managed ETLworkflows using Apache Airflow, achieving a 30% reduction in processing time and increasing data extractionrates by 25%, significantly improving pipeline efficiency.• Leveraged advanced SQL queries to extract and analyze large datasets from relational databases, providingactionable insights that drove a 10% increase in user engagement metrics for the platform.• Applied statistical techniques, including time series forecasting and regression analysis, to predict userengagement trends, enabling data-driven strategic decisions and achieving a 10% boost in user engagement.• Conducted comprehensive A/B testing experiments to refine the platform's user interface, leading to measurableimprovements in user satisfaction and a 10% increase in engagement metrics.• Designed and deployed interactive Power BI dashboards, tracking key performance indicators (KPIs). Thesedashboards enabled stakeholders to identify optimization opportunities, resulting in a 45% improvement inoperational efficiency.• Streamlined development workflows by implementing Git for version control, enhancing collaboration acrosscross-functional teams and ensuring seamless project progression. Show less

Master'S Degree, Data Analytics

B.E, Computer Science