Machine Learning Engineer

Arlo • New York, NY

9d • $180,000 - $230,000

About The Position

Arlo is rebuilding health insurance for small businesses from first principles, aiming to maximize the portion of premium dollars that go towards care. They achieve this by identifying fraud, guiding members to high-quality, lower-cost care, automating operational overhead, and eliminating unnecessary vendors. Artificial intelligence is central to their operations, used across underwriting, operations, clinical programs, and member experience to create an increasingly efficient insurer. The company is operating at a significant scale, is profitable, handles hundreds of millions in premiums, covers tens of thousands of members, and is growing rapidly. They are backed by prominent venture capital firms and have a team with experience from leading tech companies and healthcare organizations.

This role focuses on Machine Learning infrastructure for Arlo's underwriting, which is a core business function. The Machine Learning Engineer will build and manage the infrastructure for training models on extensive patient and claims data, and for serving real-time quotes with low latency against massive inference datasets. The position also involves creating tools to enhance the iteration speed of data scientists and actuaries. While the role is primarily infrastructure-focused, there is an opportunity to engage in ML and data science work, testing and evaluating personal ideas alongside supporting the data science team.

Requirements

• Strong track record building ML or data infrastructure in production at scale.
• Deep proficiency in Python.
• Comfort in processing large datasets (Spark, Databricks, or equivalent).
• Experience with model training pipelines and/or low-latency model serving in production.
• Experience building tooling that makes other people faster (feature testing, experiment tracking, backtesting, or similar developer/researcher-facing infrastructure).
• Ability to own systems end-to-end, set standards, and operate reliable production infrastructure (SLAs, monitoring, on-call).
• Genuine interest in the modeling itself, with a desire to engage in data science work beyond just infrastructure.

Nice To Haves

• Prior experience in a regulated space like healthcare or insurance.
• Experience with MLOps tooling (MLflow or similar), feature stores, or experimentation platforms.
• Experience supporting data science or actuarial teams in production environments.

Responsibilities

• Build and own the infrastructure layer for underwriting models trained on tens of millions of patients and hundreds of millions of rows of claims data.
• Ensure training is reliable, reproducible, and scalable as data volume and model complexity increase.
• Build and own the API layer for real-time quoting, serving trained models against inference-time datasets of trillions of rows.
• Own the latency, reliability, and scalability of the serving path for the quoting product.
• Make it easy for data scientists and actuaries to test new features and ideas.
• Build backtesting and validation infrastructure for quick and trustworthy model performance measurement.
• Simplify the process from idea generation to a validated, production-ready model, making experimentation easier.
• Operate reliable production infrastructure, including setting standards, monitoring, and on-call responsibilities.

Benefits

• Equity
• High ownership and responsibility from day one
• Opportunity to work on an important mission that directly influences how people access care and improves lives
• Growth and expansion opportunities with increasing scope and career velocity
• Apply AI to fundamentally reimagine healthcare
• High pace, high collaboration environment
• Competitive compensation

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed