My Road to ML
The repository was created to track my progress in Machine Learning related topics in order to organize my knowledge and goals. Includes machine learning projects and showcases in Scikit-Learn, TensorFlow, and Keras.
About
About Me
Data Engineer with a strong software development background and 3+ years of experience using predictive modeling, data engineering, and analyzing large data sets to solve challenging business problems.
Experience
Data Engineer - New York Life Insurance
Apr 2020 - Present
QA Engineer - NYCHA
Oct 2018 - Apr 2020
QA Analyst - Royal Bank of Scotland
Aug 2015 - Sep 2018
Education
Master of Science in Analytics (Expected May 2024) - Georgia Tech
Data Analytics Boot Camp - Columbia University
Bachelor of Science in Banking and Finance - University of Wales
Portfolio
Checkout few of my works
The Simpsons Characters Recognition App
Application built on Convolutional Neural Network (CNN) to recognize The Simpsons characters. Tech stack: Python, CNN, Tensorflow, Flask, JavaScript, HTML, CSS, Bootstrap.
ML Pipeline for Short-Term Rental Prices in NYC
Project on building a complete end to end ML pipeline to predict rental prices for AirBnB rentals using MLFlow and Weights and Biases.
Salary Predictor Application with FastAPI
Salary prediction model that is trained on Census Bureau data and deployed with FastAPI. A remote DVC pointing to AWS S3 bucket that tracks data changes. In addition, wrote unit tests to monitor the model performance on various slices of the data. Deployed my model using the FastAPI package and created API tests. Both the slice-validation and the API tests are incorporated into a CI/CD framework using GitHub Actions.
Belly Button Biodiversity
An interactive dashboard of belly button micro-organisms developed from a dataset of volunteer provided samples. Data is served from a Python Flask application and dynamically gathered and presented with javascript and Plotly.
US Wildfire Dashboard
The dashboard shows the number of fires that happened each year, and it also provided users the ability to see the number of fires by cause of the fire. D3 is used to make the dashboard interactive. An interactive map illustrates the occurrence of the fire by location, time, and fire size. Kepler.gl map is used. Currently, the map has only data from 2015. Tech stack: Python, Flask, CSS, HTML, Bootstrap, JavaScript, SQLAlchemy, D3, Kepler.gl.
Music-Related Venue Analysis: New York City
Research on most popular music venues in NYC using Foursquare API data. Analysis has been done on popular categories by neighborhood, top liked, and rated venues, a correlation between likes, ratings, number of venues, and neighborhoods. Tech stack: Python, Pandas, Matplotlib, API, Linear Regression, Gmap.
MatPlotLib Challenge
Visualization of Pyber and Pymaceuticals data using MatPlotLib. Data processing was done with Pandas.
Surfs Up
Basic climate analysis and data exploration of climate database using Python, SQLAlchemy, Pandas, and Matplotlib. Climate data has been take from OpenWeatherMap API.
D3 Times
This project uses D3.js to visualize some state-level data about population health based on 2014 U.S. Census data. It simulates an on-line newspaper article with an interactive visualization.
Global Earthquake Map
Visualization of earthquakes occurring globally using a live API feed provided by the USGS. The data in the feed is updated every 5 mins, and includes all earthquake data for the Past 7 Days.
UFO Sightings
Filter and look up UFO sightings data.
ETL Project
Project on building dataset of all upcoming concerts in NY state area and join MTV artist data to have more information about artists and concerts.
Weather By Latitude
An analysis of weather patterns using python, the OpenWeatherMap API, and CityPy to visualize trends in temperature, humidity, cloudiness, and windiness by latitude.
Mission to Mars
A webscraping application which retrieves and presents summary information, the latest news, and images of Mars.