Why this role existsOur models decide whether a care team walks into a room tonight. They train on 70,000 years of continuous vital signs joined to clinical records from more than 400,000 unique patients. When one of them degrades it does not surface as an error rate. It surfaces as a patient who deteriorated and nobody was alerted, which means the platform that trains, ships, and watches those models carries real clinical weight. You will own it: the pipelines, the path to production, and the monitoring that catches degradation before a clinician would.
What you'll own- Pipeline orchestration. Training, evaluation, and deployment workflows in Airflow, with automated retraining, promotion, and failure recovery.
- Deployment and release. Models onto our platform on AWS including Batch, with versioning and rollback through MLflow, maturing toward shadow and canary releases.
- Tracking and lineage. MLflow registry, conventions for artifacts and metadata, and dataset versioning so training runs are reproducible.
- Monitoring and drift. Drift, prediction quality, and degradation alerting on models where degradation is clinically consequential.
- ML compute and cost. AWS compute for training and inference, infrastructure-as-code, and cost optimization.
- Hands-on model work. Contributing to model development alongside the ML engineering team, as a secondary focus behind the platform.
- Compliance. HIPAA and SOC 2 across pipelines, with sound PHI handling in training data, artifacts, and outputs.
Required Qualifications- 4+ years in MLOps, ML engineering, DevOps, or a closely related infrastructure role
- Strong Python for pipeline development, tooling, and automation
- Hands-on Airflow, and a model registry such as MLflow
- Deploying and operating ML workloads on AWS (Batch, EC2, S3, IAM, CloudWatch)
- Containerization, infrastructure-as-code, SQL, and Snowflake
- Building monitoring and alerting for production systems
- Enough model development experience to contribute alongside ML engineers
Preferred- Model serving frameworks or data versioning tools
- Healthcare, medical devices, or clinical data systems
- Significant open source, systems that outlived your tenure, or a high-bar engineering background
$150,000 - $220,000 a year
As a full-time Senior ML Ops Engineer, you will be employed by Circadia Health, Inc. The anticipated annual base salary range for this full-time position is $150,000 - $220,000. The base range is determined by role and level, and placement within the range will depend on a number of job-related factors, including but not limited to your skills, qualifications, experience, and location.
Annual salary is only one part of an employee's total compensation package at Circadia Health. We also offer:
- Meaningful employee stock options
- 100% company-paid medical, dental, and vision coverage
- 401(k)
- Competitive time off with pay policies including vacation, sick days, and company holidays
- Impact: your work will influence care decisions for tens of thousands of seniors every day.
- Culture: hard-working, mission-driven, and collaborative - with weekly and monthly social events like yoga, beach bonfires, and Wednesday/Friday team lunches.
