MLOps Engineer

Clair Health, Inc.; Menlo Park (CA); Presencial; USD 140.000 - 230.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Competitive salary

Equity

Health, dental and vision insurance

401(k)

Paid time off and company holidays

Catered team lunches

Descripción de la vacante

Clair Health is building the first wearable for hormone-aware health, a wrist-worn band and app tracking hormones, sleep, HRV and activity. We seek an MLOps Engineer to own training, packaging, deployment, monitoring and rollbacks for models across cloud and device, ensuring reliability as software evolves.

You will ship models to mobile devices, optimize data pipelines, and maintain robust observability while scaling for thousands of members in a small Menlo Park team.

Formación

• 2 to 5 years in MLOps, ML infrastructure, platform or backend engineering.
• Strong Python, containers and orchestration, and real experience with a major cloud provider.
• You've built CI/CD pipelines and run production data pipelines or job queues.
• Reliability habits: alerts that mean something, runbooks, and post-incident fixes that stick.
• You want to work in person with a small team in Menlo Park.

Responsabilidades

• Deploy and run ML services and batch processing jobs in a containerized cloud environment.
• Own the pipeline from sensor data to stored biomarkers, including retries and backfills when a model changes.
• Keep builds and deployments fast, lean and reproducible.
• Build model versioning, reproducible training and release gates, with regression tests against known-good outputs.
• Make rollouts and rollbacks safe and boring.
• Own the path for shipping models to mobile devices, and ensure on-device and cloud results agree.
• Monitor models and data in production: input quality, drift, latency, failures and cost.
• Manage training compute and keep cloud spend in check.

Conocimientos

Python

Containers

Orchestration

CI/CD

Production data pipelines

Descripción del empleo

This is a role for someone who likes making ML systems dependable at the moment they start to matter. Every Clair band sends a steady stream of data, and a chain of models turns it into biomarkers, sleep and hormone insights. As our MLOps Engineer, you will own how those models are trained, packaged, deployed, monitored and rolled back, in the cloud and on the phone. At launch, that chain has to work every day for thousands of members. It has to keep working as models, firmware and app versions change underneath it, and you will be the one who makes sure it does.

Description

About Clair

Clair Health is building the first wearable for hormone-aware health. Clair is a wrist-worn band and app that tracks reproductive hormones noninvasively, alongside sleep, recovery, heart rate, HRV and activity. Our models are trained against hormone measurements, so women can see their cycle as it actually unfolds instead of as a calendar guess. Founding Member reservations are open and devices ship in January 2027. We're a small team in Menlo Park, backed by Khosla Ventures and a16z speedrun.

About the role

This is a role for someone who likes making ML systems dependable at the moment they start to matter. Every Clair band sends a steady stream of data, and a chain of models turns it into biomarkers, sleep and hormone insights. As our MLOps Engineer, you will own how those models are trained, packaged, deployed, monitored and rolled back, in the cloud and on the phone. At launch, that chain has to work every day for thousands of members. It has to keep working as models, firmware and app versions change underneath it, and you will be the one who makes sure it does.

What you'll do

Serving and pipelines

• Deploy and run our ML services and batch processing jobs in a containerized cloud environment.
• Own the pipeline from sensor data to stored biomarkers, including retries, backfills and reprocessing when a model changes.
• Keep builds and deployments fast, lean and reproducible.
• Build model versioning, reproducible training and release gates, including regression tests against known-good outputs.
• Make rollouts and rollbacks safe and boring.
• Own the path for shipping models to mobile devices, and make sure on-device and cloud results agree.

Monitoring, data and cost

• Monitor models and data in production: input quality, drift, latency, failures and cost.
• Run observability and evaluation for our LLM features, including cost and latency.
• Build the data infrastructure for training sets from our studies, with access controls, de-identification and retention policies suited to health data.
• Manage training compute and keep cloud spend in check.

Who you are

• 2 to 5 years in MLOps, ML infrastructure, platform or backend engineering.
• Strong Python, containers and orchestration, and real experience with a major cloud provider.
• You've built CI/CD pipelines and run production data pipelines or job queues.
• You understand ML well enough to tell a model regression from an infrastructure bug.
• Reliability habits: alerts that mean something, runbooks, and post-incident fixes that stick.
• You want to work in person with a small team in Menlo Park.

Nice to have

• Infrastructure as code.
• Running relational and analytical databases at scale.
• Deploying models on mobile devices.
• LLM observability and evaluation tooling.
• Work with health data, HIPAA, or another regulated environment.

What we offer

• Competitive salary and meaningful early equity
• Health, dental and vision insurance
• 401(k)
• Paid time off and company holidays
• Catered team lunches
• Direct work with the founders and real ownership of our ML platform