Joe Laniado

Data Scientist | Programmer

My Portfolio

Personal Skills

Get an insight on what I know and love to do.

Data Visualization

It is more than just graphs. It is about creating a story behind the data to make it easier for a person to draw insights from.

Machine Learning

There is nothing more exciting than using advanced statistical models to predict outcomes, gain insights, or automate processes. Check out some of my implementations below.

Programming

Proficient in Python, R, SQL, and more. More than 6 years combined programming experience in educational and proffesional settings.

Statistical Analysis

From describing and interpreting data to hypothesis testing, forecasting, and simulating systems; statistical analysis represents a big part of what I do.

Communication

Developed great written and verbal communication skills by practicing public speaking, group decision-making, and conflict resolution while in the military.

Business Intellingence

Learned to integrate a specific business model and KPIs in the analytical process. This allows to create better insights in defining data-driven solutions.

joe's image

Hello There!

My name is Joe Laniado and I am currently a graduate student at Georgia tech pursuing a master's degree in Analytics. Previously, I worked as a data analyst and programmer at Logic Studio, a software development company where I operated with a team to deliver data driven solutions to multiple local and international clients. During that time, I also had the opportunity to teach a couple of courses on data analytics at one of the local Universities.

I studied Computer Science at the United States Air Force Academy then went on to serve as an officer in the Panamanian military after graduating. Having grown up in Panama and attended portions of high school and all of my undergrad in the US, I am fluent in Spanish and English.

In my free time I enjoy spending time in the outdoors climbing, camping, hiking, snowboarding, and more, to truly appreciate the natural beauty of the world. I like science fiction reading, Dune being my favorite book of all time. I also have two dogs, a drama-loving rottweiler named Roxy and a giant coonhound-malamute mix named Hank. Feel free to get in touch if you want to chat or if you just want a picture of those two!

Portfolio

Bird species Identification

An exploratory implementation of computer vision by using 4 different machine learning subdisciplines to separate approaches to recognize the species of a bird from a picture. Methodologies include: dimensionality reduction, traditional statistical classification models, density estimation, and convolutional neural networks.

  • Category: Computer Vision, Data Augmentation, Data Wrangling
  • Tools: Python (Sklearn, Tensorflow, OpenCV), Jupyter Notebook, Latex
Demo image

Company Bankruptcy Prediction

A project that focuses in predicting company bankruptcy based on 95 different variables. Mostly financial metrics. It serves as an implementation of a machine learning pipeline, from exploratory data analysis and transformation to model selection and hyperparameter tuning. The data was collected from the Taiwan Economic Journal.

  • Category: Machine Learning, Exploratory Data Analysis, Statistical Inference
  • Tools: Python (Numpy, Pandas, Matplotlib, Sklearn), Jupyter Notebook
Demo image

Feature Based Water Potability Prediction

In a world where drinkable water is rapidly becoming a scarce resource, it is important to find a way to quickly identify a possible source as safe for human consumption. This project is an attempt at using machine learning classification to predict water potability based on only 10 variables. The data consists of different water quality metrics for 3276 different water bodies.

  • Category: Exploratory Data Analysis, Data Wrangling, Supervised Learning
  • Tools: Python (Random Forest, Boosting, SMOTE), Jupyter Notebook
Demo image

Wildfires in the Continental U.S.

Using data from the Forest Service Research Data Archive, a visualization was created to provide a quick insight into the wildfire history in the continental United States from 1992 to 2018. The data was stored in the form of a SQLite database file which was queried and cleaned using R in order to prepare it for creating the visualizations. The dashboard was created using tableau.

  • Category: Data Visualization, Exploratory Data Analysis, Databases
  • Tools: Tableau, SQLite, R
Demo image

Life Satisfaction VS GDP per capita

An exploratory data analysis project that tries to determine if a relationship between life satisfaction and economic development exist. The data consists of GDP per capita and life satisfaction surveys for more than 150 countries in a span of around 10 years. It serves a demonstration that simple visualizations techniques can be powerful tools in gaining important insights about a specific population.

  • Category: Data Visualization, Exploratory Data Analysis, Data Cleaning
  • Tools: Tableau, R (Misc. Tidyverse Packages)
Demo image

Image Compression using Machine Learning

A deployed example of using unsupervised learning for image compression. It decomposes an image into 3 color pixels and groups similar points together to reduce the amount of memory required to store it. The model is then deployed into a user-friendly web application to upload their own pictures, compress them, and be able to download the result into their own computers. Due to compression issues, only bmp images are supported.

  • Category: Unsupervised Learning, App Development, MLops
  • Tools: Python (Kmeans, Numpy, Opencv), Streamlit
Demo image

SafeCity

A team effort to provide an interactive crime rate visualization tool for prospective homeowners and renters in L.A. city. It allows the user to rank neighborhoods on its level of safety depending on 10 different crime categories and visualizes previous crimes that happened recently in the area. It serves as an application to provide a quick insight on different parts of a city to make better decisions when finding real state.

  • Category: Data Visualization, Data Wrangling, Statistical Analysis
  • Tools: HTML, CSS, D3.js, Google maps API
Demo image

Portfolio Website

Side project to showcase work in a user friendly and non-technical way. Good refresher for web development skills as well as an opportunity to learn how to use the Bootstrap library. Feel free to play around with it and any feedback is appreciated!

  • Category: Web development, Fun Learning Experience
  • Tools: HTML, CSS, JS, Bootstrap 5, Github Pages
portfolio website