This role pays less than
57%
of similar roles. Most pay
$142,500–$237,250
— the shaded band above.
At the midpoint, this role pays about
$169k
versus about
$190k
for comparable roles.
Based on 240 similar postings.
Employer
About Upstart
Upstart is an AI lending platform that partners with banks and credit unions to expand access to affordable credit using non-traditional variables.
Upstart currently has
73 open roles
on FindRole.
Listed pay typically runs
$166,900–$230,450
across 72 roles with salary data.
As an Applied Scientist on the Unsecured Underwriting Machine Learning team, you will research and implement enhancements to core underwriting models that inform credit decisions for unsecured lending products. You will be responsible for researching machine learning and statistical approaches to improve predictive performance, designing and evaluating model improvements through rigorous experimentation, and analyzing downstream effects to ensure reliable decisioning. Collaborating with engineering teams, you will translate research findings into production-ready solutions and documented methodologies. The role requires a graduate degree in a quantitative field and experience using Python to develop and evaluate models. Key competencies include knowledge of probability, statistics, supervised learning, feature engineering, and managing complex datasets. This position addresses the technical challenge of maintaining high research velocity while ensuring the reliability and responsible oversight required for production-level credit underwriting systems.
What does a Applied Scientist earn?
Median $208800 from 47 postings across 10 companies.
Research machine learning and statistical approaches to improve predictive performance of unsecured underwriting models.
Design, implement, and evaluate model enhancements using rigorous experimentation and validation methods.
Analyze model performance and downstream effects to ensure proposed changes improve decisioning reliably.
Partner with engineering teams to support technical reviews, implementation, and deployment of model improvements.
Translate research findings into clear recommendations, documented methodologies, and production-ready solutions.
Balance research velocity with analytical rigor and responsible oversight for production underwriting models.
What we're looking for
Graduate degree in mathematics, applied mathematics, statistics, physics, econometrics, operations research, computer science, or a related quantitative field.
0–2 years of experience conducting machine learning, statistical modeling, or applied quantitative research in an academic or industry setting.
Experience developing and evaluating machine learning or statistical models using Python.
Demonstrated knowledge of probability, statistics, and machine learning methods.
Experience designing experiments or validation analyses to assess model accuracy, reliability, or predictive performance.
Preferred PhD in a quantitative field.
Knowledge of supervised learning, model evaluation, and feature engineering techniques.
Strong written and verbal communication skills to explain technical concepts and collaborate with cross-functional teams.