Chemical Engineering Data Science

SANKARSHAN ACHARYA

******************@*****.***

***** ********* ***** 760-***-**** (Cell)

Escondido, CA 92026

HIGHLIGHTS

Strong analytical and problem-solving skills. Able to lead projects with rock-solid commitment to customer satisfaction

Experience using Python, JavaScript and Tableau for visualization and reporting

Significant experience with relational and non-relational databases. Significant experience with MapReduce, Spark and Hadoop for the analysis and storage of large datasets

Solid background in statistics and modeling methodology. Able to work with complex data and drive discovery. Knowledge of supervised and unsupervised learning

Good programming knowledge including Python, R, Java and MATLAB EDUCATION

• University of California, San Diego

Masters' in Data Science and Engineering

Capstone project involved predicting real estate rental rate using Airbnb data. Model yield: R2 of 85% and MAE of 50. Commercial potential being explored

B.S. Applied Mathematics

B.S. Chemical Engineering

• Coursera

Deeplearning.ai

UCSD Java Programming: Big Data

Johns Hopkins University: Data Science

• Udemy: Excel Analytics

SKILLS

Python: pandas, seaborn, geopandas, matplotlib, numpy, scikit-learn, statsmodels, nltk, TensorFlow, Keras, PyTorch

R: dplyr, diffEq, ReacTran, rsconnent, quanteda, RWeka, tm

PostgreSQL, Oracle, Map/Reduce, Hadoop, Spark, Hadoop, AWS

Java, Ruby on RailsBridge, Ember, Shell scripting, Node-RED

Microsoft Office: Word, Excel PowerPoint

Visualization with Tableau, Power BI

Data Files: JSON, XML, CSV, Graph Data

Knowledge of ETL, data cleaning

Proficient in Software Testing Life Cycle (STLC) methodology

Statistical modeling, Deep learning, Ensemble Methods, Time series analysis, NLP EXPERIENCE

Transcore, Data Analyst June 2019-August 2019

Worked on Tableau and SSRS based reporting

o SSRS based reporting used procedures written in SQL Server and RDLs (Report Definition Language) to take the tables created by the procedures and format them in a way the customers want

o Tableau based reporting is similar in that it uses the same procedures (and maybe tables) and uses them to create visualizations which are put together in a way as to present a Gestalt (whole picture) for the report

Initiative towards venture funding February 2019– present

Product to provide insights into the financial health of the companies for investors o Financial data (e.g. income statements, balance sheets, and cash flow statements) for publicly traded companies is graphically displayed to enable decisions o Prototyping reports using R, Power BI, Apache Tomcat web server to create a PoC. o Browser-based app to receive requests from the user for the reports is under construction. Apache Tomcat webserver is being used for the purpose Contract Programmer and Modeler April 2015 – present

Assisting clients with modeling and programming

Ormat tech is a renewable energy company. Using a combination of ARIMA and regression, predicted abnormal operating conditions. Great initial results. Work in progress

Tru Halo is an online market place for health, wellness and beauty products. Working on a product recommendation engine for use by customers

Mercury Health aims to predict falls in patient populations. Used a combination of convolutional neural network and feedforward neural networks to predict likelihood of fall. Great initial results. Work in progress

Training professionals in mathematics (statistics, calculus), programming and engineering Verisk, Data Science Intern June 2017 – September 2017 A diverse analytics company focused on risk across many domains

Developed statistics-based analytics to predict length of stay in a healthcare facility. Enabled at least 50% improvement in the accuracy of predictions

Employed linear regression and Gradient boosting techniques. Developed the modeling approach to include many sub-models in lieu of a single global model

Developed initial models for assessing energy disruption risk

Conducted a survey of the NLP methods for detecting fake news stories related to energy disruption. Created an initial model using PyTorch with an accuracy of over 90%. Model needs further refinement Betasoft, Development Trainee April 2015 – Sept 2015 Software development organization

Enabled criteria-based testing of created models for public use on a robust platform

Created Test Cases and designed test plans for smoke, sanity, functional, system, integration and unit testing. Tested Android and IOS mobile devices and applications with Selendroid, Appium, and Robotium in conjunction with Selenium

Alphaszenszor, Contract Modeler August 2014 – Feb 2015 Sensor chip design and production start up

Worked on the temperature control system for the sensor chip

Derived the model based on principles of electrical circuits and heat transfer. Solved the derived equations using Mathematica Dsolve and Manipulate

The evolved analysis represented significant improvement over published literature and was used in seeking series B funding for the company

ACTIVITIES

Member, Tau Beta Pi, National Engineering Honors Society January 2014 – present

Contact this candidate