Jeremiah Toribio

Data Scientist

Jeremiah Toribio


Data Scientist


(956) 204-6042



Available Work Locations: Anywhere • Boulder, CO • Charlotte, NC • Colorado Springs, CO • Las Vegas, NV • New York, NY • Phoenix, AZ • Remote • TX • Washington D.C.

Military Veteran

Active Top Secret

Some college

Core Technologies: Python, MySQL, Spark, Tableau, Pandas, NumPy, Matplotlib, Seaborn, SciKitLearn, Anaconda, Jupyter Notebooks, Git/GitHub

Core Competencies: Data Storytelling, Applied Statistics, Machine Learning, Natural Language Processing, Classification, Regression, Clustering, Time Series Analysis, Anomaly Detection

Primary Focus: Data Analytics

Additional Skills: Crypto Currency

Hire Me Because

A natural leader with a gravity-like pull towards motivation/keeping composure & high quality standards and values. Proven experience working on Air Force Special Operations platforms, I ebb and flow through stressful situations. Proficient data professional with a keen eye for cleaning data and utilizing machine learning models. Proudly bringing said skills to future teams along with a creative and innovative energy.

Skills Include:

  • Recognizing patterns within data
  • Effective & efficient communication
  • Intuition for taking effective action

Values Include:

  • Courage
  • Respect
  • Compassion

Best Characteristics Include:

  • Steadfast
  • Innovative
  • Confident

My Capstone Project: Art of Data

With the goal in mind to remove the human element when appraising the price of fine art. With art auction sale data retrieved from having a control of 10 artists, 100 pieces per artist. A large bulk of this data was to clean the data, which was retrieved in the form of a PDF, it required retrieving a blob of text and problem solving to retrieve a single entry that would then need regex to extract the data within the line. Create a model using regression & deep learning methodologies that can accurately predict the hammer price (final sale price) of a piece of art that will provide value to the art industry in the form of higher transparency as well as provide better estimations for auction houses like Sotheby’s and Christie’s. The best performing model was a LassoLars 90% of a median baseline 590% better than the mean baseline. This could prove useful when comparing to the 44% of our data sample, accuracy of prices that fall within art appraisal estimation. Findings could help improve the volume at which art is traded due to being able to make data driven decisions that may assist in the purchasing of art.

“With technology, I am most passionate about the avenues of problem solving it enables. It is the gateway to advancement in society, allowing for efficient solutions at the largest scale. My passion is deep rooted in curiosity; and what the product these tools will deliver and grateful that I can partake in such changes!