Sam McCormick

Data Scientist at Tagup, Inc.
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Somerville, Massachusetts, United States, US

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
You need to have a working account to view this content.

Experience

    • United States
    • Software Development
    • 1 - 100 Employee
    • Data Scientist
      • Aug 2018 - Present

      Building survival models to predict time-to-failure for industrial equipment; engineering data pipelines for training and validation of survival models at scale by leveraging distributed computing Building survival models to predict time-to-failure for industrial equipment; engineering data pipelines for training and validation of survival models at scale by leveraging distributed computing

    • United States
    • Insurance
    • 1 - 100 Employee
    • Data Science Intern
      • May 2018 - Jul 2018

      Develop models to quantify insurance risk using state-of-the-art techniques in natural language processing and deep learning; curate and analyze data on commercial inspections, accidents, and fire safety metrics; hypothesize and test factors to infer accident probability using machine learning models Develop models to quantify insurance risk using state-of-the-art techniques in natural language processing and deep learning; curate and analyze data on commercial inspections, accidents, and fire safety metrics; hypothesize and test factors to infer accident probability using machine learning models

    • United States
    • Higher Education
    • 700 & Above Employee
    • ScM, Data Science
      • Aug 2017 - May 2018

      Selected projects: Crime Prediction in San Francisco (https://github.com/sjmccorm1993/Crime-Prediction) • Using historical crime and eviction data from 2003-2015, built a model using gradient boosted trees to predict whether future crime would occur in a given census tract • Engineered a simulator that predicts and plots numbers of crimes in each census tract, with an average absolute error of just over one crime per tract when forecasting 24 hours into the future • Created a blog documenting process and summarizing results (https://medium.com/crime-prediction) Building a Search Engine • Developed a search engine that first builds an inverted index given a collection of XML documents, then given a user query returns and ranks relevant search results by a weighted combination of popular metrics such as TF-IDF and PageRank Predicting Amazon Employee Access (Kaggle challenge) • Built an ensemble classifier by combining various tree-based methods from scikit-learn in order to predict access requirements for new Amazon employees (https://medium.com/@samuel_mccormick/using-tree-based-ensemble-methods-to-predict-amazon-employee-access-1631497a5c14) Show less

    • United States
    • Higher Education
    • 1 - 100 Employee
    • Research Assistant
      • Jun 2017 - May 2018

      • Work with Professor Matthew Kraft on various projects within the field of quantitative education policy • Evaluated effectiveness of SchoolCNXT mobile app for improved parent-teacher communication using survey and raw mobile app data, culminating in a research memo sent to the New York City Department of Education for review • Created datasets containing state-level test NAEP test score means and standard deviations for a project examining teacher accountability, requiring extensive research into the complex sampling design and weighting scheme of the NAEP assessment Show less

    • United States
    • 1 - 100 Employee
    • Research Assistant/Programmer
      • Aug 2015 - Jun 2017

      Designed and implemented statistical programming algorithms to analyze, describe, and modify data in order to inform policy decisions at both the federal and state level. Developed a series of computer programs to clean and rigorously verify the validity of student data from twelve school districts nationwide as part of an evaluation of expanded learning time and data-driven instruction (DDI) in elementary education. Wrote Stata code to implement multiple imputation and impact analysis, and report findings in tables for internal review. Cleaned, categorized, and tabulated teacher interview data collected from selected teachers from several of the study schools (2016-2017) Implemented impact analyses measuring the effect of three programs put into place by the Department of Health and Human Services, Office of Adolescent Health (OAH) to delay subsequent teen pregnancies, improve contraception use, and promote school attachment and completion for expectant and parenting females on five designated outcome variables correlated with successful high school graduation (2016) Worked under the supervision of senior economic researchers to analyze the correlation between teacher evaluation scores, demographic characteristics, and student performance as measured by value-added modeling for students and teachers in Oklahoma public schools. Analyzed the effectiveness of the Relay Graduate School of Education by examining New York City public school data and comparing the achievement gains realized by students taught by graduates of the Relay program to the gains of students taught by non-Relay teachers (2015-16) Show less

    • United States
    • Education Management
    • 1 - 100 Employee
    • Research Assistant
      • Jun 2014 - Nov 2015

      Performed hands-on research for a Brown and a Yale professor working in tandem to determine the roles of bureaucrats in the implementation of anti-poverty programs in India Collected and analyzed data using Excel and Stata Used Stata to analyze and regress data in order to separate causal from non-causal effects Performed hands-on research for a Brown and a Yale professor working in tandem to determine the roles of bureaucrats in the implementation of anti-poverty programs in India Collected and analyzed data using Excel and Stata Used Stata to analyze and regress data in order to separate causal from non-causal effects

    • Hall-Hoag Project Asssistant
      • Jun 2013 - Aug 2013

      Archived Hall-Hoag Collection of Dissenting and Extremist Literature; compiled a file database using Microsoft Excel and catalogued thousands of documents; analyzed and grouped documents according to historical content Archived Hall-Hoag Collection of Dissenting and Extremist Literature; compiled a file database using Microsoft Excel and catalogued thousands of documents; analyzed and grouped documents according to historical content

    • Research Services
    • 700 & Above Employee
    • Laboratory Assistant
      • Jun 2009 - Aug 2009

      Worked closely with a Brown University neuroscience professor to determine which reward pathways of the brain were affected by morphine use Worked closely with a Brown University neuroscience professor to determine which reward pathways of the brain were affected by morphine use

Education

  • Brown University
    Master of Science - MS, Data Science
    2017 - 2018
  • Brown University
    Bachelor of Arts - BA, Applied Mathematics/Economics
    2012 - 2015
  • Barrington High School
    2007 - 2011

Community

You need to have a working account to view this content. Click here to join now