ML Platform Engineer — Scale AI Deployments

Foundry AI Partners; Northern (KY); On-site; USD 120,000 - 160,000; Full time

Job summary

Foundry AI Partners seeks a Platform Engineer to own and scale the ML infrastructure powering model deployment and serving. You will craft scalable platforms, implement robust monitoring, and optimize end-to-end workflows across client environments.

The role emphasizes building CI/CD for training and deployment, reducing latency, and managing costs on GPU/CPU workloads while ensuring security and governance for enterprise clients.

Qualifications

• 4+ years of experience in platform engineering or DevOps.
• Strong knowledge of Kubernetes, Docker, and cloud infrastructure (AWS/GCP/Azure).
• Experience with ML model serving frameworks (TorchServe, TensorFlow Serving, etc.).
• Proficiency in Python and infrastructure-as-code tools (Terraform, Pulumi).
• Understanding of ML workflows: training, evaluation, deployment, monitoring.
• Excellent debugging skills and operational mindset.

Responsibilities

• Design and maintain ML infrastructure for model deployment and serving
• Build monitoring and observability systems for production AI agents
• Implement CI/CD pipelines for model training, evaluation, and deployment
• Optimize inference latency and cost across GPU/CPU workloads
• Develop tooling for experiment tracking, versioning, and reproducibility
• Ensure security, compliance, and data governance for enterprise clients

Skills

Platform engineering

DevOps

Kubernetes

Python

Cloud infrastructure

ML model serving

Education

Bachelor's degree or equivalent in CS/Math/Engineering

Tools

Kubernetes

Docker

Terraform

Pulumi

TorchServe

TensorFlow Serving

Job description

Foundry AI Partners seeks a Platform Engineer to own and scale the ML infrastructure powering model deployment and serving. You will craft scalable platforms, implement robust monitoring, and optimize end-to-end workflows across client environments.

The role emphasizes building CI/CD for training and deployment, reducing latency, and managing costs on GPU/CPU workloads while ensuring security and governance for enterprise clients.