Sankar Panda Email & Phone Number

Hyderabad, Telangana, India

India

Led a cross-functional team in architecting and implementing real-time data processing solutions using Google CloudPub/Sub and MQTT protocols, resulting in a 40% reduction in downtime on the production line.Designed and optimized data pipelines using Google Cloud DataProc and BigQuery, enabling efficient analysis of largedatasets and reducing query times by 50%.Implemented machine learning models utilizing YOLO for defect detection and quality control, leading to a 30%improvement in product quality.Collaborated with manufacturing engineers and data scientists to identify key business challenges and develop tailoredmachine learning solutions to address them.Conducted regular training sessions for the engineering team to enhance their understanding of data engineering andmachine learning conceptsImplemented DDR4 and DDR5 memory technologies in server configurations, enhancing data processing capabilities andsupporting large-scale manufacturing data analyticConducted performance optimization and troubleshooting for RAM, DRAM, and SSD-based systems, resolving complextechnical issues to minimize production downtime.Data Technologies: BigQuery, Google Cloud DataProc, Google Cloud Pub/Sub, MQTT, Apache KafkaMachine Learning: YOLO(You Only Look Once), TensorFlow, scikit-learnProgramming Languages: Python, Java, SQLDatabases: MySQL, PostgreSQL,MongoDBTools: Apache Beam, Apache Airflow, Tableau, Power BICloud Platforms: Google Cloud Platform (GCP), AWS,AzureVersion Control: Git, GitHub

Engineered and maintained ETL pipelines using PySpark, processing large-scale data from various sources, and loading itinto AWS S3 for analytical processing, achieving 99.9% data availability.Implemented real-time data streaming solutions using Apache Kafka and Spark Streaming, enabling timely analysis andvisualization of streaming data for operational insights.Collaborated with AWS architects to optimize storage and compute resources on S3 and EMR clusters, ensuring costeffective and scalable data processing capabilities.Led training sessions for junior data engineers, providing mentorship on best practices, coding standards, and the effectiveuse of Databricks and PySpark.Implemented data governance and quality checks, establishing best practices to maintain high-quality data in the data lake,leading to a 25% improvement in data accuracyConducted performance tuning and optimization of Spark jobs, reducing processing time by 30% and lowering operationalcosts

India

Led the development of end-to-end data integration solutions using Azure Data Factory, ensuring efficient data movementand transformation across cloud and on-premises environments.Implemented data pipelines for diverse data sources, enabling seamless data flow and real-time analytics for businessintelligence purposes.Collaborated with cross-functional teams to gather requirements, design scalable ETL architectures,and optimize data workflows, resulting in a 30% improvement in data processing efficiency.Conducted performance tuning and optimization of Azure Data Factory pipelines, reducing processing time and improvingoverall system reliability.Trained almost 45 freshers on python developmentData Integration: Azure Data Factory, Apache NiFiCloud Platforms: Microsoft Azure, AWSProgramming Languages: Python,SQLMachine Learning Libraries: scikit-learn, TensorFlow, KerasDatabase Technologies: SQL Server, MongoDB,MySQLVersion Control: GitTools: Jupyter Notebook, PyCharm, Visual Studio CodeSoft Skills: Team Collaboration, ProblemSolving, Communication

Hyderabad Area, India

Project Overview: Predicting customer churn with Python: Logistic regression, decision trees and random forestsTechnology: Python, DB2, NumPy, Matplotlib, Tableau and Scikit Learn.Roles & Responsibilities:• Conduct data regression analyses of the Policy Renewal, Cancellation & Agent Termination• Have achieved 80% accuracy rate on the Random Forest and Regression Model• Have set up data pipeline for the same and emailing the prediction result to the stake holders• Modernize data streamlining processes, resulting in a 25% reduction on Policy Cancellation and Agent Termination

Hyderabad, Telangana, India

Technology: Python, DB2, NumPy, Matplotlib, Tableau and TensorFlowProject Overview: Automatic Data extraction from scanned document for faster claim process.Roles & Responsibilities:• Apply machine learning methods on large data sets and extract data through OpenCV.• Segregate the photo & Text from the document applying different of Algorithms.• Run small scale experiments with new algorithms and pass it to CNN for if it’s a genuine claim• Work with other stakeholders and prioritize data/ML initiatives across the product• Assume ownership over certain components and continuously work to improve them• Focus on user impact and how to build a better product

Hyderabad

Twitter Sentiment Analysis:Technology: Python, NumPy, Spacy, NLTK, Matplotlib,KAFKAProject Overview: Identify how an Insurance company performing In PublicRoles & Responsibilities:.Getting the twitter streaming data with the tweepy module into the sytem with JSON format.• Segregate the features from Tweet using tweepy and apply different Spacy Functions Method to get Features• Extract Sentiments from Tweets using Text blob• Applied different methods from Matplotlib to get correct visualization of sentiments.• Assume ownership over certain components and continuously work to improve them• Focus on user impact and how to build a better product

Rice Grain Classification:Technology: Python, NumPy, Matplotlib, TensorFlow & OpenCVProject Overview: Detect Rice quality Percentage in Different Rice sampleRoles & Responsibilities:Apply machine learning methods on large data sets and extract data through OpenCV.• Segregate the features from Rice Image and apply different thresholding Method to get Features• Run small scale experiments with new algorithms and pass it to CNN to identify rice quality• Work with other stakeholders and prioritize data/ML initiatives across the product• Assume ownership over certain components and continuously work to improve them• Focus on user impact and how to build a better product