AI Platform Engineering and MLOps/LLMOps Resident - Strong exp in NIM, Triton, NeMo, RAPIDS and GPU placement and CI/CD or GitOps and environment promotion

Role Overview

Platform Engineering

• Define reusable NIM and Triton deployment patterns across H200 and L40S pools.
• Create templates for model packaging, endpoints, GPU placement, health checks, logging, metrics, security controls, and rollback.
• Support NeMo-based model customization and fine-tuning on dedicated H200 capacity.
• Use RAPIDS for GPU-accelerated data preparation, embeddings, document processing, and related pipeline stages.
• Define model lifecycle processes for experimentation, registration, evaluation, approval, deployment, monitoring, rollback, and retirement.
• Establish CI/CD or GitOps patterns for promotion between development, validation, and production environments.
• Create RAG and document-intelligence onboarding patterns for ingestion, embeddings, vector indexes, access control, and rollback.

Technical Focus

• NIM, Triton, NeMo, RAPIDS, GPU placement, and MLOps/LLMOps.
• Model registries and model, prompt, and version traceability.
• Model evaluation, approval gates, monitoring, rollback, and retirement.
• RAG, embeddings, vector indexes, document processing, retrieval evaluation, endpoint observability, and security controls.

What Is Offered

• The company states that an excellent base salary, company benefits, a collaborative environment, and career progression are offered.

Requirements

• 8+ years of experience.
• Strong experience with NVIDIA NIM, Triton, NeMo, RAPIDS, and GPU placement.
• Experience with CI/CD or GitOps and environment promotion.
• Knowledge of model registries and model, prompt, and version traceability.
• Experience with model evaluation, approval gates, monitoring, rollback, and retirement.
• Experience with RAG, embeddings, vector indexes, document processing, and retrieval evaluation.
• Knowledge of endpoint observability, security controls, and service-level measures.