Ashutosh Parekh

About Me

Hi! I am currently working as a Lead Software Engineer at Salesforce, Inc. based in San Francisco primarily working with the infrastructure team on distributed systems. I hold a Masters degree in Computer Science from the University of California, San Diego with focus on Machine Learning and Artificial Intelligence.

I am interested in converting ideas to code. I dabble with various technologies to build reselient applications and software solutions. I have prior experience in competitive programming and web and app development.

Apart from these, I like playing the drums as a hobby and listening to Rock/Metal music \m/. Also, I love going on scenic hikes, be alongside nature, read interesting books and travel extensively.

Work Experience

Lead Software Engineer - Salesforce, Inc. (Feb 2019 - Current)

Work with deploying the HBase database on the public cloud using Kubernetes. Create scripts and process management code and package it using Docker along with the security related changes. Work on the algorithms for auto-scaling components and its implementation to save on the overall cost-to-serve. Responsible for creating a set of test cases, documentation for knowledge articles and ensuring the continuous delivery of a highly scalable and consistent database to be served via the public cloud.

Software Engineering Intern - Salesforce, Inc. (June 2018 - September 2018)

With the Big Data Multi-tenancy team, worked on the improvement and optimization of Apache HBase and Phoenix Query Server.

Technology Analyst - JPMorgan Chase & Co. (July 2016 - August 2017)

Worked in the Corporate Technology Line of Business. Experience with working in Java, AngularJS, C# in an Agile manner.

Developer Intern - Quantiphi Analytics Pvt. Ltd. (June 2015 - July 2015)

Data Management Tools in PHP for Quality Assurance Teams. Also worked on a News Aggregator Engine.

Projects

Music Generation using RNN Architectures

Experimented with different architectures of Recurrent Neural Networks to generate music at a character level using ABC notation. Utilized LSTM and GRU architectures with varying levels of hidden units and layers to get the best performing model which could generate respectable music. See report for more details.

View Report

Prediction of Salary Income Ranges using Supervised Learning

Built a predictive model that predicts the possible salary range of a user based on a few selected features. We used the Stack Overflow 2016 data-set which provides a myriad of features for the said task. Used various Supervised Classification Learning Algorithms like Random Forests, XGBoost, LightGBM, Neural Networks and compared the performances.

View Report

Image Captioning using Deep Neural Networks

Developed a novel architecture where an additional RNN layer was introduced in between the CNN and LSTM layers for image captioning. The advantages and disadvantages of the new architecture were analyzed in detail using various metrics like BLEU score. Further details in the report.

View Report

Multi label Document Tagging

Developed a supervised model to assign multiple relevant tags to biomedical clinical notes. The model suggests tags based on the Medical activity and condition mentioned in the document. Further details can be found in the project report inside the link.

View Project