Engineer, Platform Engineering - AI

Intercontinental Exchange (ICE); Atlanta (GA); On-site; USD 120,000 - 180,000; Full time

Job summary

Intercontinental Exchange, Inc. is seeking an experienced Infrastructure Engineer to deploy and manage GPU clusters and the ML workflow platform with AI capabilities.

You will handle driver management, CUDA toolkits, and container runtimes, while building scalable images with PyTorch and TensorFlow. The role covers observability, RAG pipelines, agent memory, and multi-agent orchestration in a 24/7 environment.

Qualifications

• Bachelor's degree preferred.
• 3+ years in infrastructure engineering, systems administration, or DevOps.
• 3+ years scripting and automation (Python, Ansible, GitOps).
• 3+ years hands-on Kubernetes in production.
• 2+ years Linux administration.
• Direct experience with GPU infrastructure (NVIDIA preferred).
• 1+ years CUDA experience.
• 1+ years MCP experience.
• 1+ years vector databases and embedding infra.
• 1+ years RAG pipeline design and deployment.
• 1+ years agent memory patterns.
• 1+ years agentic AI systems orchestration.
• 1+ years semantic search and ANN techniques.
• 1+ years workflow/orchestrion automation tools.
• Experience with enterprise monitoring and observability tools.
• Ability to work in a service-oriented team environment.
• PM, organization, time management.
• Customer focus and strong user experience.

Responsibilities

• Deploy, configure, and maintain GPU clusters and related infrastructure.
• Design and maintain the workflow automation platform with AI capabilities.
• Manage NVIDIA driver versions, CUDA toolkits, and container runtimes.
• Build container images with ML frameworks (PyTorch, TensorFlow).
• Implement monitoring, alerting, and observability for GPU infra.
• Deploy and maintain vector store infrastructure for RAG pipelines and semantic search.
• Implement end-to-end RAG workflows including ingestion, chunking, embedding, retrieval.
• Tune agent memory systems and context windows.
• Deploy, operate, and maintain Agentic AI systems and orchestration frameworks.
• Monitor MCP servers, configurations, and access controls.
• Develop automation to improve platform reliability and efficiency.
• Provide L2/L3 support and vendor escalation.
• Implement security controls including RBAC and secrets management.
• Execute change requests and maintain technical docs.
• Support production operations 24/7 and provide problem resolution guidance.
• Mentor team members and on-call staff.

Skills

English fluency

Team collaboration

Project management

Customer focus

Communication

Education

Bachelor's degree preferred

Tools

Kubernetes

Linux

NVIDIA GPUs

CUDA

Python

Ansible

GitHub Actions

NVIDIA drivers

Vector databases

RAG pipelines

MCPs

Job description

Job Purpose

We are on a mission as a team. We are problem solvers and partners, always starting with our customers to solve their challenges and create opportunities. Our start‑up roots keep us nimble, flexible, and moving fast. We take ownership and make decisions. We all work for one company and work together to drive growth across the business. We engage in robust debates to find the best path, and then we move forward as one team. We take pride in what we do, acting with integrity and passion, so that our customers can perform better. We are experts and enthusiasts - combining ever-expanding knowledge with leading technology to consistently deliver results, solutions and opportunities for our customers and stakeholders. Every day we work toward transforming global markets.

Responsibilities

• Deploy, configure, and maintain GPU clusters and associated infrastructure
• Designing, building, and maintaining the workflow automation platform that uses AI capabilities (LLM/MCP/Agentic capabilities)
• Manage NVIDIA driver versions, CUDA toolkits, and container runtimes
• Build and maintain approved container images with ML frameworks (PyTorch, TensorFlow, etc.)
• Implement monitoring, alerting, and observability for GPU infrastructure
• Deploy and maintain vector store infrastructure for RAG pipelines, agent memory, and semantic search
• Implement and maintain end-to‑end RAG workflows, including document ingestion, chunking, embedding generation, and retrieval optimization
• Maintain and tune agent memory systems, including short‑term context windows, long‑term persistent memory stores, and episodic memory retrieval patterns
• Deploy, operate, and maintain Agentic AI systems, including multi‑agent orchestration frameworks and tool‑use pipelines
• Deploy, host, and maintain MCP servers within the containerized platform infrastructure
• Manage MCP server configurations, versioning, access controls, and integration with agentic workflows
• Monitor MCP server health, performance, and availability; respond to incidents and perform root cause analysis
• Develop automation and tooling to improve platform reliability and efficiency
• Provide L2/L3 technical support and vendor escalation for complex issues
• Implement security controls including network policies, RBAC, and secrets management
• Execute change requests and maintain technical documentation
• Respond to and assist in production operations in a 24/7 environment
• Provide technical analysis, resolve problems, and propose solutions
• Provide support to, and coordinate with, developers, operations staff, release engineers, and end‑users
• Educate and mentor team members and operations staff
• Participate in a weekly on‑call rotation for after‑hours support

Knowledge and Experience

• Bachelor's degree preferred.
• 3+ years in infrastructure engineering, systems administration, or DevOps
• 3+ years in scripting and automation skills (Python, Ansible, GitOps)
• 3+ years hands‑on experience with Kubernetes in production
• 2+ years experience with Linux administration
• Direct experience with GPU infrastructure (NVIDIA preferred)
• 1+ years experience using CUDA
• 1+ years experience using MCPs
• 1+ years experience with vector databases and embedding infrastructure
• 1+ years experience with RAG pipeline design and deployment
• 1+ years experience with agent memory patterns (in‑context, external stores, retrieval‑augmented memory)
• 1+ years experience with agentic AI systems using orchestration frameworks
• 1+ years experience with semantic search, embedding models, and ANN search techniques
• 1+ years working with workflow/orchestrion automation tools
• Experience with enterprise monitoring and observability tools
• Ability to work in a service‑oriented team environment
• Project Management, organization, and time management
• Customer focused, and dedicated to the best possible user experience
• Communicate effectively with both technical and business resources
• Fluent speaking, reading, and writing in English

Desired Knowledge and Experience

• 1+ years of experience with AI developer toolkits (NVIDIA drivers, CUDA, cuDNN, and NCCL)
• 1+ years of experience with Run:AI, NVIDIA AI Enterprise, or DGX systems
• 1+ years of experience with n8n
• 1+ years of experience with GitHub Actions

#LI-SH3

#LI-ONSITE

Intercontinental Exchange, Inc. is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to legally protected characteristics.