AI Platform Engineer — GPU & Vector AI Infra

ICE; Jacksonville (FL); On-site; USD 120,000 - 180,000; Full time

Job summary

Intercontinental Exchange, Inc. seeks an AI Platform Engineer to deploy and optimize GPU clusters, vector stores, and MCP servers within a containerized platform. You will enable AI-driven workflows across LLM, MCP, and Agentic capabilities, ensuring secure, scalable operations.

The role requires deep expertise in GPU infrastructure, Kubernetes, and automation, with a focus on reliability, observability, and cross-team collaboration in a fast-paced environment.

Qualifications

• Bachelor's degree preferred.
• 3+ years infrastructure engineering, systems administration, or DevOps.
• 3+ years scripting and automation: Python, Ansible, GitOps.
• 3+ years Kubernetes in production.
• 2+ years Linux administration.
• 1+ years GPU infrastructure (NVIDIA preferred).
• 1+ years CUDA usage.
• 1+ years MCPs usage.
• 1+ years vector databases and embedding infrastructure.
• 1+ years RAG pipeline design and deployment.
• 1+ years agent memory patterns (in-context, external stores, retrieval memory).
• 1+ years agentic AI orchestration frameworks.
• 1+ years semantic search and ANN.
• 1+ years workflow/orchestrion automation tools.
• Fluent English: speaking, reading and writing.

Responsibilities

• Deploy, configure, and maintain GPU clusters and infrastructure.
• Design and maintain the AI workflow automation platform (LLM/MCP/Agentic).
• Manage NVIDIA drivers, CUDA toolkits, and container runtimes.
• Build container images with ML frameworks (PyTorch, TensorFlow).
• Implement monitoring, alerting, and observability for GPU infra.
• Deploy vector store infra for RAG pipelines, memory and search.
• End-to-end RAG workflows: ingestion, chunking, embedding, retrieval.
• Tune agent memory systems (context windows, long-term memory).
• Deploy and maintain Agentic AI systems with multi-agent orchestration.
• Host and maintain MCP servers within containerized platform.
• Manage MCP configurations, versioning, access controls.
• Monitor MCP health, incidents, root-cause analysis.
• Develop automation to improve platform reliability.
• Provide L2/L3 support and vendor escalation.
• Implement security controls (RBAC, secrets, network policies).
• Execute change requests and maintain technical docs.
• Respond to production ops in a 24/7 environment.
• Provide technical analysis and propose solutions.
• Coordinate with developers, operations, release engineers, end-users.
• Educate and mentor team members; participate in weekly on-call.

Skills

English fluency

3+ years DevOps

Python

Ansible

GitOps

Kubernetes

Linux admin

GPU infrastructure

Vector databases

RAG pipelines

Agent memory

Orchestration frameworks

Semantic search

Workflow automation

Education

Bachelor's degree

Tools

Kubernetes

GitHub Actions

Run:AI

NVIDIA AI Enterprise

NVIDIA CUDA

cuDNN

NCCL

n8n

Job description

Intercontinental Exchange, Inc. seeks an AI Platform Engineer to deploy and optimize GPU clusters, vector stores, and MCP servers within a containerized platform. You will enable AI-driven workflows across LLM, MCP, and Agentic capabilities, ensuring secure, scalable operations.

The role requires deep expertise in GPU infrastructure, Kubernetes, and automation, with a focus on reliability, observability, and cross-team collaboration in a fast-paced environment.