Senior AI Systems Engineer: Agentic Retrieval Pipelines

NVIDIA Corporation; Santa Clara (CA); On-site; USD 184,000 - 357,000; Full time

Job summary

NVIDIA is seeking a System Software Engineer - Agentic Retrieval to design and optimize GPU-accelerated data processing pipelines for multi-modal content and LLM workflows. You will collaborate with LLM researchers to build full-stack, scalable tools for preparation, deduplication, filtering, and classification of training corpora, aiming for high accuracy and low cost.

The role focuses on Rust and Python components, benchmarks, profiling, and integrating with containerized CI/CD pipelines.

Qualifications

• Bachelor’s or Master’s Degree in CS, CE, or related field or equivalent experience.
• 6+ years of professional experience in a similar role.
• Cloud-based software delivery experience and cloud infra patterns.
• Knowledge of MLOps: Docker-Compose, containers, Kubernetes, data-center deployments.
• Strong hands-on NLP/LLM/VLM/Generative AI and Agentic Retrieval workflows.
• Self-starter with growth mindset and strong written/spoken communication.
• Ability to collaborate across multi-functional teams and geographies.

Responsibilities

• Develop and optimize Rust-based data processing frameworks for large datasets on GPU-accelerated environments.
• Lead Agentic Retrieval pipeline components with GPU acceleration to improve TCO.
• Collaborate with LLM/ML researchers to build full-stack, GPU-accelerated data prep pipelines for multimodal models.
• Implement benchmarking, profiling, and optimization of algorithms in Python across architectures.
• Coordinate with teams to deliver production-ready tools and libraries within the LLM ecosystem.
• Contribute to CI/CD pipelines and provide peer reviews on performance, scalability, and correctness.

Skills

NLP/LLM

Agentic Retrieval

Cloud architecture

Performance optimization

Cross-functional collaboration

Education

Bachelor's / Master’s in CS or related

Tools

Rust

Python

Docker-Compose

Kubernetes

Job description

NVIDIA is seeking a System Software Engineer - Agentic Retrieval to design and optimize GPU-accelerated data processing pipelines for multi-modal content and LLM workflows. You will collaborate with LLM researchers to build full-stack, scalable tools for preparation, deduplication, filtering, and classification of training corpora, aiming for high accuracy and low cost.

The role focuses on Rust and Python components, benchmarks, profiling, and integrating with containerized CI/CD pipelines.