AI Safety Platform Engineer: Reproducible, Secure Experiments

NVIDIA; New York (NY); On-site; USD 152,000 - 241,500; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking a Harness/Platform Engineer in New York to help build the foundation that makes experiments reproducible and trustworthy. You will design, implement, and maintain the agent harness and evaluation infrastructure, ensuring runs are repeatable from environment to result and that tooling remains versioned.

Collaborate with security research and evaluation engineers; contribute to NeMo-based tooling and workflows.

Qualifications

• Bachelor's degree (or equivalent) with 5+ years of software engineering experience.
• Strong Python engineering plus comfort with containerized environments and CI/CD.
• Experience with agent frameworks, LLM orchestration, ML infrastructure, or evaluation harnesses.
• Reproducibility instincts: care for versioned tooling and repeatable results.

Responsibilities

• Harness development: Design, implement, and maintain the agent harness.
• Evaluation infrastructure: Build the systems we use to run and reproduce experiments.
• Reproducibility: Own environments, tooling, and repeatable runs.
• Collaboration: Partner with security and evaluation engineers on their workflows.

Skills

Python

Agent frameworks

CI/CD

Education

Bachelor's degree or equivalent

Tools

Docker

Kubernetes

NVIDIA NeMo

Job description

NVIDIA is seeking a Harness/Platform Engineer in New York to help build the foundation that makes experiments reproducible and trustworthy. You will design, implement, and maintain the agent harness and evaluation infrastructure, ensuring runs are repeatable from environment to result and that tooling remains versioned.

Collaborate with security research and evaluation engineers; contribute to NeMo-based tooling and workflows.