GOKUL KRISHNASWAMY

adgu9n@r.postjobfree.com 720-***-**** linkedin.com/in/gokul-krishnaswamy Edison, NJ github.com/GokulKrishnaswamy public.tableau.com/profile/gokul.krishnaswamy PROFESSIONAL SUMMARY

Data Science professional with 2 years of experience designing and implementing solutions. Recent graduate with strong analytical skills and demonstrated leadership experience. Area of specialization includes machine learning, data analysis and algorithms.

SKILLS

Languages : SQL, Python, R, Java, Hive, C++, XML, HTML, JSON Database : Oracle, MySQL, PostgreSQL, MongoDB

Tools : Apache Spark, Databricks, Git, Tableau, Oracle SQL Developer, AWS, MS Office Packages : Scikit-learn, NumPy, Pandas, SciPy, Keras, Matplotlib, Seaborn, PySpark Certifications : AWS Machine learning speciality, AWS Cloud practitioner EDUCATION

MS, Data Science University of Denver Denver, CO, USA GPA: 3.86 out of 4 August 2018 – August 2020

BTech, Computer Science and Engineering Amrita Vishwa Vidyapeetham University Coimbatore, TN, India GPA: 8.57 out of 10 August 2014 - July 2018

RELEVANT COURSES

Python, Database (SQL), Machine learning, Data Visualization, Data Structures and Algorithms, Big Data, Statistics and Probability, Data mining, Parallel and distributed computing, Natural language processing. EXPERIENCE

Data and Python Developer intern Hitachi ABB Power grids Broomfield, CO, USA May 2020 – August 2020

• Developed scripts in python to extract data from websites. Deployed scripts to production.

• Rewrote a python script to merge large number of datasets and optimized its runtime by factor of 3.

• Improved logistic regression model’s performance (F1 score) by using effective EDA and data manipulation in python.

• Collaborated with data analysts to write ETL jobs (with views and stored procedures) in SQL.

• Conducted code reviews and debugged scripts before production in agile environment. Data and Research Coordinator University of Denver Denver, CO, USA July 2019 – May 2020

• Head of a diverse team of 15 members. Increased operational efficiency of the research team by 60%.

• Redesigned the architecture of the project and reduced its complexity. Integrated and archived 1TB volume of data.

• Generated qualitative data by observing over 500 research videos, cleaned data and derived statistical measures from it.

• Designed solutions in AWS using S3 and machine learning services to eliminate manual work, worth 150 work hours. Data Analyst/Scientist Onceptual Business Solutions Coimbatore, TN, India July 2018 – December 2019.

• Designed over 20 normalized tables in Oracle. Cleansed data and migrated them with ETL pipelines in python.

• Visually analysed data by creating Tableau dashboards. Created over 100 ad hoc SQL queries to validate the results.

• Created an analytical model(clustering) and identified patterns in data which improved customer count by 30%.

• Created an association rule mining algorithm to identify items that are frequently bought together.

• Presented data reports to clients and recommended business changes which improved profit by 40%. Software Developer intern Oracle Bangalore, KA, India January 2018 - June 2018

• Developed over 60 automation test scripts in python and deployed them in linux server. Performed unit testing. RELEVANT ACADEMIC PROJECTS

Sensor data processing: Created a solution in AWS platform to collect, store and analyse sensor data. Created backend using AWS Lambda, designed REST API through API Gateway and stored data in a NoSQL database (DynamoDB) using Boto3. Insurance cost prediction: Performed exploratory data analysis, feature engineering and data manipulation to create LASSO regression model to predict insurance cost based on patient medical records. Achieved a record low RMSE value on test data. Churn modelling: Created a customer churn model in python using artificial neural network. Used ROC curve to find ideal trade-off value. Optimized the model through hyper-parameter tuning. Achieved accuracy of 85% and recall score of 94%. Simple image classification: Performed dimensionality reduction and developed a logistic regression model using TensorFlow library that classifies the handwritten digit to a numerical value. Achieved accuracy of 91% on test data. Bacteria cases prediction: Forecasted the number of bacteria cases in Melbourne city (Time series, real world data) by developing a seasonal ARIMA model in python. After optimization, achieved a record low RMSE value. RESEARCH PAPERS PUBLISHED

K. Gokul, R. Pooja, K. Gowtham and G. Jeyakumar, "A self-switching base vector selection mechanism for differential mutation of differential evolution algorithm," 2017 International Conference on Communication and Signal Processing

(ICCSP), Chennai, 2017, pp. 1545-1549, doi: 10.1109/ICCSP.2017.8286647.

• Developed a strategy (RandtoBest) to optimize Differential Evolution Algorithm for uni-modal functions. Presented the paper at an IEEE conference.

K. Gokul, R. Pooja and G. Jeyakumar, "Empirical Evidences to Validate the Performance of Self-Switching Base Vector Based Mutation of Differential Evolution Algorithm," 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, 2018, pp. 2213-2218, doi: 10.1109/ICACCI.2018.8554928.

• Studied the critical parameters and provided mathematical evidences to validate the performance of RandtoBest strategy.

Contact this candidate