Machine Learning Engineer

Neal Vaghasia

Moorpark, CA ***** 805-***-****

************@*****.*** www.linkedin.com/in/nealv/ github.com/NV1 SKILLS

TECHNICAL SKILLS

Python • SQL • Javascript

• Java • MATLAB • LA

TEX

Libraries/Technologies

Keras • PyTorch • Apache Spark

• Flask • Pandas • NumPy

• React/Redux • Visual Studio • Git

• DialogFlow • Juypter • Scikit-Learn

• MongoDB • Matplotlib • Spacy

• NLTK • Pivotal Cloud • Anaconda

• Docker • Hadoop

SOFT SKILLS

Languages

• English • Spanish • Gujarati

EDUCATION

UNIVERSITY OF CALIFORNIA

SAN DIEGO

BACHELOR OF SCIENCE

MACHINE LEARNING AND NEURAL

COMPUTATION

Dec 2019 La Jolla, CA

TEXAS TECH UNIVERSITY

MASTER OF SCIENCE

COMPUTER SCIENCE

June 2021 Lubbock, TX

COURSEWORK

BACHELOR

Data Structures + Algorithms

Object-Oriented Programming in Java

Artificial Intelligence Algorithms

Natural Language Processing +

Practicum

Modeling and Data Analysis

Data Science + Practicum

Probability

Brain Computer Interfaces with Deep

Learning

Statistical Analysis

MATLAB + Practicicum

Linear Algebra

Vector Calculus

Calculus Engineering I + II

Analytic Geometry

Design for Development

WORK EXPERIENCE

CIRCUIT EXPRESS, INC

MACHINE LEARNING ENGINEER

Nov 2017 - Present Simi Valley, CA

• Developing PCB Defect python automated system with FastAi and TetryonAi with high accuracy digit recognition to detect costly errors 85% quicker than human eye for rocket PCBs deployed to SpaceX Falcon 9

• Integrating and building classification, regression, and pattern recognition models with consumer PCB sales SQL relational database to efficiently increase prediction ratings to 80% for the expense categories that 105+ clients can immediately reduce

• Researching and experimenting with Tensorflow and PyTorch on currently deployed projects with the R&D department to reduce running time by 15% and increase output by 8%

• Collaborating with engineering staff to configure and design software for customer relationship management and workflowmanagementsystemsonTFS TECHNICAL PROJECTS

COVID19 AI CHATBOT DEMO, GITHUB

April 2020 - Present Los Angeles, CA

• Designed and built with Google DialogFlog and Flask an end to end bot that is able to answer 70% queries/FAQ related to COVID-19 on Telegram UI

• Utilized RAPID API to show real-time case statistics of worldwide, country-wise, Indian State, and District-wise areas

• Experimented and ran local tests on POSTMAN for configuration and debugging for quick deployment on the Pivotal Cloud

• Implemented a saved log database of user and Bot conversations with MondoDB Atlas cluster using AWS

LSTM STOCK PORTFOLIO OPTIMIZER GITHUB

May 2019 - Present Moorpark, CA

• Constructed an momentum indicator LSTM Neural Net model for an optimal asset portfolio collection with Tensorflow Keras, NumPy, and normalization equations to predict time series steps and sequences with S&P 500 data from 2000 - 2018

• Implemented Kernel Density Estimation and K-Means clustering algorithms for cross-validation, pattern matching, and accuracy testing on years 2013-2016 collected from the Investor’s Exchange API

FAKEDBOOK WEBSITE, GITHUB

March 2019 - Present La Jolla, CA

• Built an end to end minimal clone of Facebook with Ruby on Rails for REST API to construct a life cycle methodology for custom social media sites

• Implemented BCrypt for a 100% secure user authentication system with password encryption

• Utilized AWS S3 for efficient access of images throughout the application and MVC architecture to keep code DRY and clean

ROBINHOOD RSI TRADER GITHUB

January 2018 - Present Moorpark, CA

• Constructed a Python based fully automated trading bot system by scraping live stock data from Robinhood API onto NumPy and Tulipy

• Added support and resistance lines to RSI to minimize risk and improve input accuracy to increase profitability from 0.04% to 1.00% per share

• Applied a scheduler to perform continuous checks by minute to minute basis on new price changes for input values to the RSI momentum indicator

Contact this candidate