VENKATESH BABU SEKAR

+1-617-***-**** ac7jor@r.postjobfree.com https://www.linkedin.com/in/venkateshbabus EDUCATION

Northeastern University, Boston, MA

Master of Science in Information Systems (GPA 3.5) Dec 2018 Data mining engineering graduate certificate Aug 2017 Anna University, Chennai, India

Bachelor of Engineering in Electronic and Instrumentation May 2013 TECHNICAL SKILLS

Data Skills: Machine Learning, Object Detection, Neural Network, Classification, Clustering Python Packages: Tensor-flow, Keras, Theano, NLTK, NumPy, Pandas, SciPy, Scikit Learn AWS: S3, Redshift, Athena, Elasticsearch Service, AWS Lambda, DynamoDB, kinesis firehose Business Intelligence: Kibana, Tableau, Quick Sight, Microsoft Power BI, Qlik Sense Languages: Python, R, SAS, Java, SQL

Database: MySQL, Microsoft SQL Server, PostgreSQL, Oracle 11g Data Integration: Talend Enterprise Data Integration, SQL Server Integration Services (SSIS) PROFESSIONAL EXPERIENCE

YouTube Channel - Science of Data: https://www.youtube.com/channel/UCStUloKik-m-Th2tPfSe6oA Jun 2018 - Present

Managing and operating a YouTube account with 60+ subscribers focused on explaining machine learning to amateurs

Edited and produced a video series (54 videos) -Introduction to Python, guiding How to start coding in Python

Developing a video series- Data science & Machine learning using Python, which covers various ML models in detail Amazon, Boston, MA - Data Analyst Co-op Jan 2018 - Aug 2018

Built Object detection classifier using fast_rcnn to detect the reverts in the pods deployed across the global FC’s

Created 500+ labelled image data set using LabelImg and generated TFRecord file to train the object detection classifier

Integrated Alexa echo with MySQL using python to provide voice-based interaction to get FC deployment details

Utilized Internet of Things platform, using Raspi to track the pod-built productivity data in real time using Quick Sight

Built live dashboard in Kibana, Tableau to track the ticket status and the Fulfilment Centre (FC) status across the globe

Developed data pipeline for efficient data flow from IoT core to Redshift, S3 and Elastic search services

Created Python script to gather data from various sources and performed data transformation using AWS lambda Northeastern University, Boston, MA - Graduate Teaching Assistant (Advances in Data Science) Sep 2017 - Dec 2017

Supervised graduate students to build various neural network models using Keras for their academic projects

Mentored students to learn the concepts of web scraping, data cleaning and data visualization using R and Python Cognizant, Chennai, India - Data Analyst (Role: Data Scientist) Sep 2013 - Jun 2016

Built a Regression model to estimate the utility of customers investment towards the organization profit

Improved prediction accuracy of model by 2% by data imputation with the logical historical values

Led a team of three associates to maintain the data dictionary and to prepare the daily and weekly reports using Tableau

Collected data using SQL scripts, cleaned and analyzed the data using Python to identify important metrics ACADEMIC PROJECTS Project Repository: https://github.com/venkateshbabusekar Security Alert - Gun Detector (Object Detection, Tensorflow, Raspi) Oct 2018

Built Object detection classifier using Raspi and Movidius Computing stick to detect the guns carried in the public gathering

Installed Tensorflow, OpenCV in Raspi and performed Object detection locally and triggered security alert email

Implemented data annotation and generated TFRecord file to train the model, deployed the model in Raspi 3 Classification of Video for Action Recognition - (Keras, Tensorboard, CNN – MLP, CNN – LSTM, LRCN) Dec 2017

Analyzed human activities and ongoing events successfully by obtaining the probabilities of the activity for the whole video

Converted all the videos into frames and extracted the feature of each image and combine then into a sequence of feature

Built various model and tuned the hyper parameters to obtain higher accuracy and visualized the output in Tensorboard Sentiment Analysis on StockTwits and News Headlines - (Keras, CNN, RNN, LSTM, MLP ) Oct 2017

Created a word dictionary from stocktwits and yahoo news headlines, tokenized and padded the words using Keras

Built a Ginsum model and calculated the vectorized distance between words, used bokeh to visualize the word distances

Projected each word in 300 dimension, built various models and tuned the hyper parameters to obtain sentiment score Enron Scandal Text Analytics (Python, Text Analytics) Jan 2017

Gathered 5 million emails related to Enron scandal and processed those data using python to implement text analytics

Identified the top 10 people who might be involved in insider trading and securities fraud which led to Enron scandal

Contact this candidate