Kelsey M.

Denver, CO, USA

Fractional Data Scientist

Professionally, I am a data scientist with skills across a broad range of methodological and engineering competencies - data mining, automation, machine learning, visualization, cloud computing, algorithmic bias, and statistics. Personally, I am a creative mind driven by the opportunity to learn.

I love creative solution-building from messy problem sets and helping organizations get started using data in a more efficient, effective way. I am passionate about streamlining and scaling data pipelines, as well as creating robust equity-first algorithms. I am particularly excited about working with non-profits and small groups led by a mission for social good.

Hire Kelsey

Work

Diversity of Economics Seminar Speakers
- Data Visualization
- R Language
A main way academic economists build networks and share and get feedback on their work is by giving seminars at other departments. Departmental choices about whom to invite to give seminars can therefore shape individuals’ careers and may contribute to or help reduce existing disparities in the profession. I supported building this dashboard, which provides information about who gives invited talks, in aggregate, by department, and for each seminar series within a department.
econseminardiversity.shinyapps.io
Consultant for Scaled Risk Models funded by Google.Org
The Student Success Tool is a product funded by Google.org and meant to scale student risk models to hundreds of institutions. I advised on and implemented scalable architecture, code, algorithms, performance metrics, and bias evaluation outputs such that data scientists can deliver unique insights and robust models to partner schools in a low-touch way.
nytimes.com
Personal GitHub page
github.com
Finalist: Centers for Medicare & Medicaid Services (CMS) AI Challenge
The CMS Artificial Intelligence (AI) Health Outcomes Challenge was an opportunity for innovators to demonstrate how AI tools – such as deep learning and neural networks – can be used to accelerate development of AI solutions for predicting patient health outcomes for Medicare beneficiaries for potential use in CMS Innovation Center innovative payment and service delivery models.

I led a team of 5 d…
cms.gov
Open Source Contribution: EvalML
I contributed a machine learning measurement, sensitivity at low alert rates, to an open source automated machine learning package.
github.com
Grand Prize Winner: Data Visualization Challenge
The AHRQ Visualization of Social Determinants of Health Challenge invited participants to develop new online tools to present and encourage use of free, publicly available social determinants of health data to better understand and predict communities’ unmet healthcare needs.

I led a team of data scientists and researchers to build a submission to this challenge, which ended up winning the grand prize!
communityconnector.mathematica.org

Experience

DataKind
Senior Data Science Consultant
Jan 2024 - Oct 2024
- Proposed plan to scale and deploy student risk model, resulting in $8 million of funding from Google.org
- Built modeling pipelines, evaluation systems, and responsible AI utilities in Azure Databricks, with a focus on student safety and supporting the most vulnerable populations
- Mentored and collaborated with data scientists and leadership on methodological and tooling choices, as well as guided technical debt priorities and best practices on code quality
Panorama Education
Senior Data Scientist
Nov 2021 - Oct 2023
- Led scoping and implementation of internal data platform, using PySpark, terraform, and Amazon Web Services (AWS), to create a scalable, practical, and secure system enabling data science, research, and product analytics work for the compan
- Established equity-first hiring and leveling processes for both individual contributors and data science managers
Mathematica Policy Research
Senior Data Scientist
Apr 2019 - Oct 2021
- Led grand-prize-winning team in Agency for Healthcare Research and Quality’s (AHRQ) data visualization challenge
- Led modeling team selected as a finalist in the Centers for Medicare and Medicaid Services’ (CMS) AI Challenge
- Developed deep learning methodology to predict hospitalizations and deaths with Google Cloud Platform (GCP), Luigi pipelines, adjustments for algorithmic bias, and an internal Python package
- Developed a scraping pipeline to feed data into an AWS database, complete with unit, integration, and end-to-end testing, as well as continuous integration (CI) functionality
uAspire
Manager of Research & Evaluation
Jul 2018 - Apr 2019
- Measured organization’s impact and predicted college enrollment of students utilizing randomized control trial data, propensity matching, and machine learning modeling
- Automated data processing through custom-built tools, including an internal R package and user interface
- Supervised analysts and steered team in expanding toolkit towards a broader use of R and Git
84.51/dunnhumby (Kroger)
Senior Data Scientist
May 2013 - Apr 2018
- Developed internal dynamic query generator to pull transactional data on over 55 million households and trained fellow analysts to use in their daily work
- Leveraged NLP methodologies to generate and personalize digital content for America’s largest supermarket
- Implemented variable reduction and unsupervised learning to cluster stores into pricing groups
Miami University
Student
Aug 2010 - May 2014
- B.A. in Mathematics
- B.S. in Quantitative Economics
- M.A. in Applied Economics