Software Engineer - MLOps

R37 Lab, R1 RCM•New York, NY

1d•$140,000 - $300,000•Onsite

About The Position

You’ll own the production runtime for Phare’s ML stack - deploying, serving, and scaling models across inference endpoints and batch/streaming workflows. You’ll build progressive delivery pipelines with automated rollouts and rollbacks, manage SLOs for latency and availability, and instrument end-to-end observability (metrics, logs, traces, drift, regression). You’ll harden the platform with Terraform, Kubernetes, and CI/CD, ensuring reproducible, auditable ML releases. We are hiring across several seniority levels ranging from Mid-level up to Staff. At a minimum, we expect 5 years of software engineering experience with 2 years of ML Ops experience. This is an in-person role in NYC, requiring at least 3 days in the SoHo office.

Requirements

5 years of software engineering experience.
2 years of ML Ops experience.
Background in operating ML systems at scale.
Experience deploying and operating models running on GPUs in production - APIs and batch/streaming inference.
Strong with Docker/Kubernetes.
Experience with IaaC (e.g., Terraform).
Experience with CI/CD for services and model artifacts.
Experience maintaining environment parity, reproducible releases, and robust model/experiment versioning with data lineage.
Experience using progressive delivery with automated rollouts/rollbacks.
Experience building end-to-end observability (metrics, logs, traces, and model telemetry for drift/regression) plus actionable alerting, runbooks, and incident response.

Nice To Haves

Experience in regulated environments (e.g., healthcare, finance).

Responsibilities

Deploying, serving, and scaling models across inference endpoints and batch/streaming workflows.
Build progressive delivery pipelines with automated rollouts and rollbacks.
Manage SLOs for latency and availability.
Instrument end-to-end observability (metrics, logs, traces, drift, regression).
Harden the platform with Terraform, Kubernetes, and CI/CD, ensuring reproducible, auditable ML releases.
Manage model registries and stage gates.
Design scheduled or event-driven retraining when appropriate.
Enforce RBAC, secrets management, encryption, and audit logs.