Senior Data Scientist

About the role

Most of what makes American healthcare expensive isn’t medical care. It’s the machinery wrapped around it: middlemen taking a cut, fraud nobody stops, and billing systems designed to fight over payment instead of deliver care. The result is higher premiums, denied claims, surprise bills, and a system patients increasingly experience as adversarial.

Arlo is rebuilding health insurance for small businesses from first principles: making sure as much of every premium dollar as possible goes to care instead of getting absorbed by the system around it. We do that by identifying fraud earlier, steering members toward higher-quality and lower-cost care, automating operational overhead, and eliminating vendors whose business exists mostly to take a cut.

AI is the foundation that makes this work. We use it across underwriting, operations, clinical programs, and member experience to build an insurer that becomes more efficient as the technology improves.

We’re already operating at meaningful scale: profitable, hundreds of millions in premiums, tens of thousands of members covered, and growing quickly through brokers, employers, and partners. Backed by Upfront Ventures, 8VC, and General Catalyst, with a team from Palantir, YC companies, and longtime healthcare operators.

About the role

Underwriting is at the core of Arlo. Every group we price depends on how accurately we can estimate the risk of the individual members inside, and the quality of the estimate is essential to the sustainability of our business. We’re hiring a Senior Data Scientist to own our underwriting model and continuously deploy measurable improvements to it based on learnings from real-world outcomes.

You’ll work with billions of claims across tens of millions of patients to identify what signals in claims history predict future medical cost, how to roll it up to a competitive price for a group, and how to deploy the system at scale.

You’ll constantly monitor the lifecycle of predictions, group policies sold, and claims incurred by our tens of thousands of members to gather novel insights that can improve our model and pricing approach.

This is a hands-on modeling role in which you will sit on the underwriting team alongside our team of data scientists and actuaries thinking through issues beyond point estimates of cost including how to handle data blindness, variability, and when the risk is too high to issue a quote. You’ll own the model but work alongside ML engineers to ensure that your ideas can be tested and deployed at scale.

What you’ll work on

Evaluate the existing Arlo underwriting model and find where it breaks

• Develop a robust evaluation framework to stress the model outcomes and identify specific gaps in risk estimates (cohorts, conditions, or claims patterns) and the underlying causal factors
• Use those findings to generate a roadmap of model and feature work that is prioritized based on making sure our rates are competitive in the market while ensuring we can remain a profitable business.

Build features that capture the full risk of a member

• Account for training and inference dataset bias to optimize member predictions
• Improve handling of member cost variance in our quoting pipeline

Experiment with model designs

• Implement different ML architectures that balance efficacy, generalizability, and understanding so we can outperform the market

Prove the lift before it ships

• Work with our backtesting harness to measure the effect of every model change on MLR and competitiveness.
• Set the bar for what "better" means and hold changes to it, so improvements to the model are trustworthy

What we’re looking for

• 5+ years as a data scientist building predictive models that made it into production
• Deep proficiency in Python and SQL, with comfort processing large datasets using Spark and using common modeling packages.
• A track record of owning a problem end-to-end in an ambiguous environment and shipping without being handed a spec
• Direct experience with healthcare data
• Strong instincts for feature engineering and model validation, and the judgment of how well results might be able to hold in production
• Interest in the larger business context. This is not a research endeavor, but a live model that directly changes our win rates, book size, and performance.
• Clear communicator who can explain what they did and why it mattered

Nice to Have

• Familiarity with claims data and their known quirks and biases
• Prior experience working at Series C or earlier start-ups
• Background in underwriting, actuarial sciences, or risk adjustments
• Experience with ML engineering and infrastructure