Tory Borsboom-Hanson, Developer in Seattle, WA, United States
Tory is available for hire
Hire Tory

Tory Borsboom-Hanson

Verified Expert  in Engineering

Data Scientist and Forecasting Developer

Location
Seattle, WA, United States
Toptal Member Since
October 31, 2022

Tory is a data scientist and forecasting specialist with extensive experience in experimental design, AB testing, and leadership roles. His stack of choice includes Python, SQL, Pandas, PySpark, Hadoop, MLFlow, XGBoost, and Scitkit-learn. Beyond data science and model development, he is an expert at data visualization and reporting, especially to non-technical audiences. He often remarks that "a poorly presented analysis can often be worse than no analysis at all."

StatisticsA/B TestingMachine LearningTime Series AnalysisPythonPandasScikit-learnSQLComputer ScienceData EngineeringQuantitative FinanceNumPyGitHubObject-oriented Programming (OOP)Amazon Web Services (AWS)Systems ImplementationXGBoost

Portfolio

Fractal
Python, SQL, PySpark, Hadoop, A/B Testing, Experimental Design...
MeridaLabs
Python 3, Experimental Design, Research, Experimental Research, Management...
MeridaLabs
Python, SQL, Pandas, NumPy, Scikit-learn, Amazon Web Services (AWS), Statistics...

Experience

Python - 6 yearsScikit-learn - 5 yearsNumPy - 5 yearsSQL - 5 yearsA/B Testing - 5 yearsPandas - 5 yearsAmazon Web Services (AWS) - 3 yearsPySpark - 1 year

Availability

Part-time

Preferred Environment

Pandas, NumPy, Scikit-learn, XGBoost, A/B Testing, Amazon Web Services (AWS), SQL, Time Series Analysis, Experimental Design, Python

The most amazing...

...tool I've built was a forecasting model currently used to guide customer engagements at a Fortune 100 bank in the US.

Work Experience

Data Science Consultant

2022 - PRESENT
Fractal
  • Assisted a Fortune 100 company in developing a time‑series forecasting model in Python using XGBoost, capable of guiding customer interactions. The model I developed provided a 9% increase in accuracy over the incumbent solution.
  • Performed A/B testing and decision science to develop high‑value business strategies and sales funnels. These strategies resulted in a 6% average increase in profit among the various platforms tested.
  • Analyzed large-scale data sets with sizes of over 900 million rows with Spark and Hadoop.
Technologies: Python, SQL, PySpark, Hadoop, A/B Testing, Experimental Design, Amazon Web Services (AWS), Pandas, NumPy, XGBoost, Predictive Modeling, ARIMA, SARIMA, Forecasting, ARIMA Models, Machine Learning, Quantitative Finance, Finance, Data Engineering, Scripting, Business Analysis, APIs

Research Team Lead

2021 - 2022
MeridaLabs
  • Designed experimental procedures for validating the economic viability of fuel-cell technologies. This resulted in the development of a novel high-temperature fuel-cell design.
  • Spearheaded the fuel-cell technology branch of the research group. I was in charge of project management, experimental direction, hiring, and people management.
  • Guided the research publication strategy for the team I oversaw. In my time here, I managed to publish eight distinct works.
Technologies: Python 3, Experimental Design, Research, Experimental Research, Management, People Management

Lead Data Scientist

2021 - 2022
MeridaLabs
  • Invented a means of using time‑series data and machine learning to predict fuel cell failures, reducing downtime and maintenance costs. This is estimated to provide over $1 million in savings over the lifetime of a large-scale fuel cell plant.
  • Performed economic modeling to estimate the cost of several varieties of hydrogen infrastructure based on scenario inputs. Using this model to optimize hydrogen infrastructure can result in a 30% cost reduction on the end user price of hydrogen.
  • Maintained AWS infrastructure and implemented data modeling of databases for use in a data warehouse. This reduced database size by 40% and doubled the average query speed.
Technologies: Python, SQL, Pandas, NumPy, Scikit-learn, Amazon Web Services (AWS), Statistics, Experimental Design, Time Series Analysis, Predictive Modeling, ARIMA, SARIMA, Forecasting, ARIMA Models, Machine Learning, Data Engineering, Scripting, APIs

IT Contractor

2007 - 2021
TBHInnovation
  • Deployed IT solutions rapidly to accommodate 100+ users in a strict high-security environment with more than 95% uptime.
  • Led teams of more than 15 people for rapid IT deployment projects.
  • Coordinated with businesses to determine IT infrastructure and security requirements for deployment at scale.
Technologies: IT, System Implementation, Deployment, IT Consulting

Techno-economics of Sub- and Supercritical Water Electrolysis

https://www.sciencedirect.com/science/article/abs/pii/S0196890422005374
Water electrolysis provides a pathway to meet increasing hydrogen demand. However, it is currently more expensive than hydrogen production methods relying on fossil fuels. By providing kinetic and thermodynamic benefits, high temperatures and pressures can increase the overall energy efficiency and reduce costs.

Point-to-point Transportation: The Economics of Hydrogen Export

https://www.sciencedirect.com/science/article/abs/pii/S036031992203107X
Hydrogen transport over long distances is a critical cost component and can involve many complex pathways. I developed a model and an associated framework that can be used to determine the cost of transport methods for both land and land-and-sea scenarios. The model assesses the transportation of liquid and gaseous hydrogen by truck, rail, barge, and gaseous hydrogen pipelines.

Carbon-neutral Fuels: Economic Analysis of Renewable Syngas Pathways via CO2 Electrolysis

https://www.sciencedirect.com/science/article/abs/pii/S0196890421006282
Producing syngas, a blend of CO and H2, is the starting point for the large-scale production of valuable chemicals, including fuels and methanol. This study evaluates pathways for syngas production from air-captured and tail-gas-captured CO2 in terms of net CO2 emissions, energy efficiency, and associated costs.

Evaluating Classification Models Using Expected Value

A guide to using the Expected Value framework to determine the value a classification model is expected to provide a business. This project also includes instructions on how to best utilize the Expected Value framework to tune the model to provide the maximum possible value.

Languages

SQL, Python, Python 3

Libraries/APIs

Pandas, Scikit-learn, XGBoost, NumPy, PySpark

Paradigms

Data Science, Object-oriented Programming (OOP), Management

Other

A/B Testing, Time Series Analysis, Experimental Design, Statistics, Experimental Research, Modeling, Classification, IT, System Implementation, ARIMA, SARIMA, Forecasting, ARIMA Models, Machine Learning, Scripting, APIs, Computer Science, Optimization, Process Economics, Value Analysis, Predictive Modeling, Quantitative Finance, Finance, Data Engineering, Business Analysis, Azure Databricks, Deployment, IT Consulting, Research, People Management

Tools

GitHub, MATLAB

Platforms

Amazon Web Services (AWS), Azure

Frameworks

Hadoop

2016 - 2020

Ph.D. in Physical Chemistry

University of Victoria - Victoria, British Columbia, Canada

2012 - 2016

Bachelor's Degree in Chemistry

University of Victoria - Victoria, British Columbia, Canada

APRIL 2022 - PRESENT

Microsoft Azure Data Scientist Associate Test Prep

Coursera

APRIL 2022 - PRESENT

Prepare for DP-100: Data Science on Microsoft Azure Exam

Coursera

APRIL 2022 - PRESENT

Perform Data Science with Azure Databricks

Coursera

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring