Bhavana Vangala

Experience

• 

  Kartha Techno Logistics Private Limited

  May 2016 - Aug 2017

  Project: www.vdelver.in Description: Ecommerce website which facilitates the customer by sending anything to anyone, by picking up something important for customers. Through this website, from eCommerce, Optical retail store, Groceries, Retail Store, a Designer Boutique, to Hyper Local Delivery Service. Name anything, it can serve them all. Responsibilities - Developed frontend in HTML5, CSS3 - Developed backend business logic in Java - Implemented ODBC for interacting with database - Implemented validations using JavaScript - Persisted data was done using MongoDB Show less

  • Business Development Executive Intern

    Mar 2017 - Aug 2017

  • Software Development Intern

    May 2016 - Feb 2017
• 

  International Journal of Computer Applications

  Jan 2017 - Nov 2017

  Publisher

  “Comparison of Decision Tree Classifier and Bayes Classifier Using WEKA”, 18 Nov 2017, International Journal of Computer Applications, Reference ID:2017915569

• 

  University of California, Riverside

  Jan 2019 - Mar 2020

  RESEARCH PROJECT IN BIG DATA MANAGEMENT Title: Open Street Map (OSM) Datasets Extraction and Visualization Objective: This research project extracts datasets and visualizes them on Open street map Role: Graduate Student Research Assistant under the guidance of Professor Ahmed Eldawy Project - Extraction and visualization of Datasets on Open Street Maps. - Developed a Pig script and a Java code to extract datasets from planet.osm data and visualized them on OSM. - Developed a Java code in Spark & Hadoop to extract datasets from osm.pbf files, joined them using relational databases, deployed it on different clusters (hdfs), AWS (EMR), and compared the efficiency of different osm file formats like .xml, .bz2 and .pbf files. Show less

  • Graduate Student Researcher

    Nov 2018 - Mar 2020

  • Graduate Teaching Assistant

    Jan 2019 - Mar 2019
• 

  Druva

  Jun 2019 - Aug 2019

  Machine Learning Engineering Intern

  Project: a) Built a predictive model to minimize the operational risk such as ensuring the customer that there are enough credits to continue backup by showing future storage usage over a 6-month horizon and credit balance over the same time horizon. b) Built a model that predicts the revenue by showing the future storage under Druva management over a 12-month horizon and the behavior of the average customer in-terms of storage consumption over a 12-month horizon. Responsibilities: Retrieved the time-series(historic) data from the Snowflake repository using SQL, python-SQL. Processed the data using python into different forms as various models take input’s different forms and tested the data on cloud (AWS). Built the above stated two models using different machine learning techniques such as Regression, Auto-ARIMA & ARIMA, Facebook’s Prophet, Amazon Forecast, LSTM’s based on large time-series data of all customers in R & Python. Studied the behavior of metrics of each algorithm and selected one that best suits the customer data. Visualization of the Prediction is done using Interactive plots in python and in R as well. Show less

• 

  Cisco

  Jul 2020 - Jan 2022

  Data Engineer Consultant

  • Conceptualized, designed, developed, and productized Contact-Hub (customer contact data) that other cisco product teams leveraged to engage and improve User experience and customer base. • Built new ETL data pipelines integrating various multi-dimensional data sources in Python (NumPy, pandas) and Spark using GCP and snowflake achieving 75% reduction in data processing times.• Created and incorporated python modules, SQL scripts, indexes and complex queries for data analysis, extraction of data sources and to automate all manual workloads involved in the ETL pipelines.• Developed a machine learning modelling framework for standardizing attributes in python which showed a potential of 60% improvement of the contactability data. • Performed ETL on real-time streaming disparate data sources to provide customer insights. Show less

• 

  Syngenta

  Jan 2022 - now

  Data Engineer

  • Designed, developed, and productized the ETL pipelines on the seeds data – phenotype and genotype that leveraged all farmers, scientists, and researchers to choose right seed for right season and crop.• Built these ETL pipelines using Python, Spark and incorporated SQL scripts in the pipeline from ingestion to the enrichment on AWS – S3, EC2, Redshift and Snowflake. Denodo is the data virtualization tool we use to connect backend database with the business teams to query multiple tables from multiple databases.• Closely work with the product teams and report metrics about the seed and its characteristics based on which the decision making is done on which seed should go in a particular season.• Using Dataiku, I integrated MLOps with the ETL pipelines in improving the quality of the data we get around the world.• Interacted with the different API’s to attain data and integrated the data to add business value to the organization. Show less