Aakash Chotrani

SDET - II

Amazon•  September 2019 - Present

• Developed core test framework for testing Lumberyard game engine. Automating regression test cases for making the Lumberyard game engine more robust which improves the end customer experience. • Created automated jenkins jobs for deploying the tools and automating the scheduled test jobs to run during build. Send failure notifications when the job fails. • Improved the framework to easily connect and test on playstation, xbox and nintendo switch.

Software Developer

Crelate Talent•  July 2018 - July 2019

• Built semantic search feature for searching candidates based on their job titles and skills by applying NLP and creating a machine learning model trained on 500K resumes which contains vocabulary of 195K unique words. Search also supports 22K job title synonyms and 2K skill search. It resulted in an increased volume of search by 5% • Created candidate similarity feature based on resume document vectors to compare similarity between candidates and to recommend the top 10 candidates similar to a given candidate. • Improved the document import process by making it asynchronous using azure async job queue. The pre-processing and parsing were made async to prevent CPU spiking on the webserver. • Implemented job description keyword extractor by implementing RAKE algorithm. • Designed LSTM network to improve resume parsing by creating a hybrid approach of using regular expressions and deep learning confidence levels to fetch unstructured data. • Researched new concepts and technologies, built prototypes, created and deliver professional presentations designed to educate all staff members and critical stakeholders.

Machine Learning Instructor

IdTech Camps•  June 2018 - July 2018

• Instructed a class of 9 students. Teaching fundamentals of Python to acquire, preprocess and train models for generating predictions. • Taught supervised, unsupervised learning methods and deep learning library Keras on top of Tensorflow to create neural networks. Mathematical and data manipulation libraries such as Scikit-learn, Pandas, Numpy and data visualization library Matplotlib. • Image recognition and classification using MNIST, CIFAR 10 and CIFAR 100 dataset. • Performed sentiment analysis using Naive Bayes theorem. Used Open Ai gym for reinforcement learning.

Research Engineer

DigiPen Research & Development•  April 2017 - May 2018

• Researched and implemented a face recognition system for the institute. Developed a network program for setting up the server and supporting multiple clients for capturing video/image feed from multiple cameras. Developed a parser for performing data analysis. • Created course recommendation & grade prediction application. • Assisted in time series prediction for Andretti Autosports for developing future Indy car race strategies. Built models and performed feature engineering. Used Tableau dashboards for data visualization.