AISHWARYA MANOHARAN
716-***-**** *****************@*****.***
www.linkedin.com/in/aishwarya-manoharan-366b7716a
https://github.com/AishwaryaMano
EDUCATION
Master of Science, Computer Science & Engineering
University at Buffalo, State University of New York, USA May 2020 Bachelor of Engineering, Computer Science & Engineering Panimalar Engineering College affiliated to Anna University, Chennai, India May 2018 TECHNICAL SKILLS
Programming Languages: Python, R, Java, CPP, C#
Frontend Technology: HTML, CSS, Java script, PHP
Data Base: SQL, MYSQL
ML Skills: KNN, Random Forest, PCA, K-means, Linear Regression, Hadoop map-reduce, PySpark Other skills: AWS Glue, Athena, CloudTrail, Tableau, JIRA, GitLab, Databricks, Azure DevOps Frameworks: Hadoop, Apache Spark, Keras, Numpy, Matplotlib, Pandas, Java Swing, Sklearns, xUnit.net CERTIFICATIONS
AWS certified developer associate, by Amazon Jan 2020 (Valid for 3 years)
Data scientist’s toolbox certified, by Johns Hopkins University on Coursera Jan 2020 (Valid for lifetime)
Mastering AWS Glue, QuickSight, Athena & Redshift Specturm on Udemy July 2020 (Valid for lifetime) WORK EXPERIENCE
Microsoft Corporation WA, USA
Software Engineer Python, Azure Data Explorer, Data Factory, DevOps, VS, SQL, Kusto Query, MS Excel 07/06/2021-Present
Took responsibility to extract signals from Bing result page and used these signals to monitor features using metric movement.
Built, maintained and automated ETL pipelines based on the above metrics. Created and owned the related dashboards.
Improved Bing image answer user experience by triggering the images based on the position at which they were triggered. Improved triggering at TOP position by 30%. Automated the pipeline to maintain the freshness of such whitelisted queries.
Performed A/B testing for above mentioned incremented image triggering and analyzed and reported the metrics movement.
Analyzed user engagement signals with the Image Answer on the Search Engine Result Page, compared metrics on page impressions with and without image answer to set criteria for triggering images and prepared a whitelist of queries for IA.
Created plug-ins and tested these plug-ins using unit testcases for ad hoc business request from partner teams. Synapsec LLC Client CVS MA, USA
Data Engineer Azure, Datarbricks, Gitlab, RALLY, Pycharm, SQL 02/15/2021-06/30/2021
Took charge of a COVID-19 dashboard. Applying the necessary filters using PySpark and SQL to calculate new metrics on daily basis, as required by the business. Maintaining the ETL pipeline on COVID-19 data. Neustar Inc. Virtual, USA
Security Team Intern AWS Glue, Athena, Cloudtrail, Redshift Spectrum, Python 07/13/2020 - 09/11/2020
Built and maintained an analytical solution using native AWS services that will create ETL transformations and data processing platform to bring efficiency in querying historical data for AWS Cloud security forensic analysis and reporting. Infosys limited Karnataka, India
System Engineer Intern PHP, MYSQL, HTML, CSS ( XAMPP platform ) 01/08/2018 - 05/02/2018 Worked on a job scheduling project where a web site was developed to view the qualifications of a chef and book them.
Was responsible for User Interface and backend for the view chef page, rating page, feedback page, blacklist page and payment.
Developed custom automation test cases to the test the application. ACADEMIC PROJECT
Single Image Super Resolution GAN Python, Keras, Scipy, Google Colab, GAN Created a model to recover finer texture details when we resolve a low-resolution image to a high-resolution image. The model was a generator discriminator model.
Concepts implemented: Residual network, Inception network, Deep neural network. Result obtained: Generated super resolution images that closely resembled the original high resolution images. Big data processing with Hadoop Python, Hadoop, Map Reduce, Pandas, Numpy Implemented the basic text processing tasks from scratch on Hadoop framework. The tasks included word count, producing modified tri-grams around keywords, producing inverted index for given data set, performing relational join on 2 tables. The obtained results were verified against the results obtained by other batches working on the same task. It was found to be similar for all the tasks implemented.
Predictive analysis Python, Matplotlib, Google Colab Designed and implemented the predictive analysis algorithms from the scratch without the use of Scikit-Learn libraries. The algorithms implemented were KNN, Random Forest, PCA and K-means. Additionally, implemented the confusion matrix, calculated the precision, and recall without the use of inbuilt functions. A desirable accuracy of 98% was achieved for random forest, 83% was achieved for K-means and 85% accuracy was achieved for KNN.