Senior AI Platform Engineer — Training & Inference Leader

ICE; Atlanta (GA); On-site; USD 180,000 - 240,000; Full time

Job summary

Intercontinental Exchange, Inc. (ICE) is hiring a full-time AI Platform Developer to architect and manage an enterprise-wide platform for AI model training, deployment, and inference at scale.

You will lead as a Senior Developer within the AI Center of Excellence, guiding MLOps and platform engineering teams and collaborating with data scientists and business stakeholders in Atlanta. The role requires deep expertise in AI/ML training pipelines, model serving, and real-time inference, with strong

Qualifications

• Extensive experience designing AI/ML training and inference platforms on cloud infrastructure.
• Deep expertise in model serving and deployment pipelines (TensorFlow Serving, TorchServe, Kubeflow).
• Proficiency with GPU clusters, distributed training, and performance optimization.
• Experience building and leading MLOps pipelines, model versioning, and CI for ML models.

Responsibilities

• Architect and manage enterprise-wide AI training and inference platform infrastructure.
• Drive innovation, operations excellence, and scalability for AI model serving environments.
• Provide technical leadership for AI platform development and cross-functional alignment.

Skills

AI platform

MLOps leadership

Kubernetes

Python

Distributed training

GPU clusters

Model serving

Education

PhD or MS/BS in CS/ML/Data Eng

Tools

TensorFlow Serving

TorchServe

MLflow

Kubeflow

Docker

Kubernetes

PyTorch

Job description

Intercontinental Exchange, Inc. (ICE) is hiring a full-time AI Platform Developer to architect and manage an enterprise-wide platform for AI model training, deployment, and inference at scale.

You will lead as a Senior Developer within the AI Center of Excellence, guiding MLOps and platform engineering teams and collaborating with data scientists and business stakeholders in Atlanta. The role requires deep expertise in AI/ML training pipelines, model serving, and real-time inference, with strong