Machine Learning Platform Engineer Full Time NEW

Software Engineer, Machine Learning Platform (ML Ops)

New York City (Hybrid)

This team of engineers is building the infrastructure that powers machine learning at scale for a Digital Media platform. Greenfield opportunity – they’re mid-build on their ML platform and looking for someone who wants to own the full stack: from model training pipelines to production inference serving, from experimentation tooling to the reliability guarantees their engineering partners depend on.

What you’ll own:

• Inference model hosting — design and operate low-latency, highly-available model serving infrastructure that engineering teams call at scale with guaranteed SLAs
• ML platform build-out — own the toolchain: model registry, feature store, experiment tracking, and serving layer — and drive its evolution as their ML footprint grows
• CI/CD for ML — build and automate end-to-end pipelines covering training, validation, deployment, and retraining
• Production reliability — implement monitoring for model health, data drift, and performance degradation; be the person who knows when something’s wrong before anyone else does
• Data scientist partnership — bridge the gap between experimentation and production, translating model requirements into scalable, containerized services
• Platform standards — set the bar for reproducibility, versioning, testing, and governance across all ML systems

What you’ll bring:

• 3+ years in software or platform engineering with hands-on ML infrastructure experience
• Strong Python or Go; comfortable at the intersection of software engineering and MLOps
• Experience building and owning CI/CD pipelines (GitHub Actions, Jenkins, or similar)
• Solid containerization and orchestration fundamentals — Docker, Kubernetes
• Cloud platform experience (AWS or GCP) and infrastructure-as-code (Terraform, CloudFormation)
• Familiarity with ML tooling: MLflow, Kubeflow, Ray, Spark, or equivalents

Bonus Points

• Experience with low-latency NoSQL datastores (Bigtable, DynamoDB)
• Observability chops — Prometheus, Grafana, Datadog, or ELK
• Data pipeline and orchestration experience (Airflow, Prefect)

This role sits at the centre of a platform that directly shapes how millions of people discover and engage with content every day. The infrastructure you build will be foundational — and the decisions you make now will define how they scale ML for years to come.