Vaibhav Rathore Email and Phone Number

• Assisted in developing ETL pipelines using Azure Data Factory, contributing to a 10% performance improvement.• Developed SQL queries in Databricks notebook for data ingestion.• Acquired data from the EPIC system using HTTP requests to a specified server path and ingested into Azure Data Lake Storage.• Managed two jobs running daily every 4 hours; directed data ingestion to current or historical jobs based on data date.• Working on cost optimization of our ADF, Databricks and Storage resources.• Currently, working on creating an ingestion framework for pipelines to acquire data from API calls.

Designed and developed a predictive analytics system to forecast household gas consumption and adjust billing based on usage. The system aims to promote energy efficiency and ensure fair billing practices for users. Key components of the project included: Data Collection and Preprocessing: • Integrated IoT-enabled gas meters to continuously collect household gas consumption data. • Aggregated additional data such as historical consumption patterns, weather conditions, and household demographics. • Applied data cleaning and preprocessing techniques to handle missing values and outliers.Feature Engineering: • Extracted and engineered features including daily and monthly consumption trends, temperature variations, and household occupancy. • Utilised feature selection methods to identify the most significant predictors of gas consumption.Model Development: • Developed a predictive model using machine learning algorithms such as Linear Regression and Random Forest to forecast monthly gas consumption per unit. • Optimized model performance through hyperparameter tuning and cross-validation.Deployment and Monitoring: • Deployed the system to ensure scalability and reliability. Set up continuous monitoring and logging to track system performance and model accuracy. • Applied additional charges for consumption exceeding set monthly limits. • Developed a user-friendly interface for customers to monitor their consumption and billing details in real-time.

• Implemented a project with the help of Python and Machine Learning algorithm i.e., K-nearest Neighbor.• Made the dataset using permutation and combination by collecting different tag names, required for training and testing the model.• Courses will be recommended based on tags provided by the user.• Technology used: Python, Machine Learning Algorithms.• Tools used: Jupyter notebook, Git.