Cell: +91-911******* E-mail: email@example.com Hyderabad, India
Programming experience in R and Python (using SCIKIT-LEARN, NUMPY, PANDAS, MATPLOTLIB, KERAS, TENSORFLOW, OPENCV, NLTK) in JUPYTER Notebook/SPYDER.
Theoretical and practical knowledge of Machine Learning techniques and their usage in classification, regression, time-series forecasting and clustering.
Experience in solving NLP problem using NLTK coupled with machine learning algorithms.
Theoretical and practical knowledge of popular Artificial Intelligence Architectures like MLP, CNN, RNN.
Ability to pick insights in the data through plots using GGPLOT2, MATPLOTLIB, TABLEAU.
Ability to pick important features and demonstrate their importance in the model for customer satisfaction.
Have a good hands-on on image processing using OPENCV library
Have good knowledge on Big Data and its tools&ecosystems
An enthusiastic intellectual and result oriented person with extensive technical abilities, inter-personal skills, analytical ability and decision-making skills. Career Profile
Company : TEKBIZ SOFT SOLUTIONS
Project Agriculture crop analysis
Client Government of Andhra Pradesh
Team Size 10
Role Data Analyst
Duration July 2016 - June 2017
The main motto of the project is to enhance the agriculture industry with IOT Technology and machine learning by recommending the different needs to the user such as weather forecasting, soil fertility levels, pesticides recommendations, crop recommendations. o Collecting the IOT Sensors data Through IBM Watson which the data denotes different attributes like Date, Time, Soil PH Level, Moisture, Temperature and more. o Had did Exploratory data analysis in excel and generated the visualization reports. o Applied Time Series and regression models in R like ARIMA to forecast the climate conditions, pest recommendation, soil fertility levels.
Project Loan Approval Prediction based on Machine Learning Approach Client TATA CONSULTANCY SERVICES
Team Size 15
Role Data Scientist
Duration July 2017 – March 2018
Predicting the Loan status whether the chosen applicant is the deserving right applicant out of all applicants for the US-based bank (Bank of America). o Developed a classification model on given historical data for predicting the status of Loan for both improvement and degradation.
o Built a classification model on given historical data to analyze the status of the loan considering the Loan ID, Applicant Income and Loan amount applied for, deliver insights to increase processing of the application and product/service development. o Applied classification algorithms such as Random Forest, Decision Trees, and various libraries. Project Term policy subscription
Client TATA CONSULTANCY SERVICES
Team Size 15
Role Data Scientist
Duration April 2018 – Present
The platform acquires data from direct marketing campaigns(phone calls) of the bank. The classification goal is to predict whether the client will subscribe to the term policy or not. o Performed a descriptive analysis to gain basic insights into the dataset, summary stasis, data cleaning, analyzing features highly impacting the target, data visualization, correlation among variables.
o Applied classification algorithms such as Logistic regression, Random Forest and libraries like numpy, pandas.
Products Under Development
Product Object Detection Using Convolution Neural Networks The project speaks of how to train a intelligent machine using AI technology where it can differentiate number of objects as per the exposure it gained by the given data, Object detection will play a crucial role in many sectors like military, hospitals, banking, retail and more to help in automation application. o Collected Data of different pictures of a particular object and marked boxes and labeling around the region of interest using Tensor Flow API which creates XML files. o Converting XML files to Excel format and start training the train files to SSD model. o After saving the trained model into a pickle file using OpenCV library in python deployed the pickle file into the video for detecting the trained objects. Product Auto Document Field Extractor Using Name Entity Recognition The application is to eliminate the manual data entry by human which helps for faster processing in sectors like Banking, Hospitals, logistics billing by using Name Entity Recognition the document fields which needs to updated in data sheets manually are updated automatically by application if it gets the pdf format of the documents, This could save both time and human error. o Had trained an empty NER model through Recurrent neural network by feeding the documents fields in Json data and trained the model.
o Deployed the saved model into Django webapp, Thus the user can upload the docs in the app and automatically fields get extracted and the labels are recognized by the name entity recognition . Academic Qualification
B.Tech (Electronics And communication Engineering) - 2016 July o Institution: Vel Tech Dr.RR & SR Technical University o Completed : 2016 JULY
o Percentage: 68%
Post Graduation Program (Big Data Analytics And Optimization) o Institution: INSOFE
o Completed : 2017 Nov
Date of Birth: 21.05.1994
Marital Status: Single
Passport Number: J9420254
Reference: Available on request.