Summary
What you’ll impact
Arlo is seeking a Senior ML Infrastructure Engineer to build and own the large‑scale training and real‑time inference infrastructure that powers its underwriting models. The role involves developing reliable, scalable systems, tooling for data scientists and actuaries, and collaborating on model experimentation.
Responsibilities
What you'll do
- Build and own the infrastructure layer that powers our underwriting model, trained on tens of millions of patients and hundreds of millions of rows of claims data.
- Make training reliable, reproducible, and scalable as data volume and model complexity grow.
- Build and own the API layer that produces quotes in seconds — serving a trained model against a much larger inference-time dataset, on the order of trillions of rows of claims across hundreds of millions of people.
- Own the latency, reliability, and scalability of the serving path the quoting product depends on.
- Make it as easy as possible for data scientists and actuaries to test new features and ideas.
- Build backtesting and validation infrastructure so model performance can be measured quickly and trustworthily.
- Remove friction from the path between an idea and a validated, production-ready model — make experimentation simpler than it's ever been.
- Build the tooling that lets our data scientists and actuaries iterate faster than ever.
- You’ll own the platform, but you'll also have the opportunity to work alongside our data scientists and actuaries to test and evaluate your own ideas — not only support theirs.
Requirements
What you’ll bring
- A strong track record building ML or data infrastructure in production at scale.
- Deep proficiency in Python, with 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.
- The ability to own systems end-to-end, set standards, and operate reliable production infrastructure (SLAs, monitoring, on-call).
- Genuine interest in the modeling itself — you want to occasionally get your hands into the data science, not only the infrastructure.
- 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.