Engineer, Platform Engineering - AI
Intercontinental Exchange20 hours ago
Atlanta, GA, USAMid Level
H1B sponsor
Responsibilities
- Deploy, configure, and maintain GPU clusters and associated infrastructure.
- Build and maintain container images with ML frameworks such as PyTorch and TensorFlow.
- Manage NVIDIA drivers, CUDA toolkits, container runtimes, vector stores, MCP servers, and agentic AI systems.
- Implement end-to-end RAG workflows, including document ingestion, chunking, embedding generation, and retrieval optimization.
- Maintain and tune agent memory systems, multi-agent orchestration frameworks, and tool-use pipelines.
- Develop platform automation, monitoring, alerting, observability, and reliability improvements.
- Implement network policies, RBAC, secrets management, and other security controls.
- Provide L2/L3 technical support, vendor escalation, incident response, root cause analysis, and production operations.
- Coordinate with developers, operations staff, release engineers, and end users while maintaining technical documentation.
- Mentor team members and operations staff and participate in a weekly after-hours on-call rotation.
Requirements
- Bachelor's degree is preferred.
- At least 3 years of experience in infrastructure engineering, systems administration, or DevOps.
- At least 3 years of scripting and automation experience with Python and Ansible.
- At least 3 years of hands-on Kubernetes experience in production.
- At least 2 years of Linux administration experience.
- Direct experience with GPU infrastructure, preferably NVIDIA, and at least 1 year of CUDA experience.
- At least 1 year of experience with MCPs, vector databases, embedding infrastructure, RAG pipelines, agent memory patterns, agentic AI systems, semantic search, embedding models, and ANN search.
- At least 1 year of experience with workflow automation tools, with desired experience in n8n.
- Experience with enterprise monitoring and observability tools, GitHub Actions, and AI developer toolkits is desired.
- Experience with Run:AI, NVIDIA AI Enterprise, or DGX systems is desired.
- Ability to communicate effectively with technical and business stakeholders and work in a service-oriented team environment.
- Fluency in spoken, written, and read English is required.
Benefits
- The position is onsite, as indicated by the job posting.
- The role operates in a 24/7 environment and includes a weekly after-hours on-call rotation.
Tech Stack
Categories
About Intercontinental Exchange
Intercontinental Exchange operates global exchanges and clearing houses, offers fixed income and commodities trading platforms, and sells market data, listings, and mortgage technology to financial institutions and lenders. Its revenue comes from transaction and clearing fees, data subscriptions, listings, and software. Founded in 2000 and headquartered in Atlanta, it owns the New York Stock Exchange and ICE Mortgage Technology, and is publicly traded on the NYSE under the ticker ICE.