Senior ML Safety Evaluator & Benchmark Architect (Equity)

NVIDIA Corporation; California; On-site; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA Corporation is seeking a Machine Learning Engineer / AI Researcher / Data Scientist to join the AI Safety and Responsibility efforts within Enterprise Risk Management. You will create robust benchmarks, run rigorous experiments, and build tools to evaluate safety of multimodal models, contributing to training methods and safeguards with reproducible results.

You will design benchmarks, define evaluation metrics, and implement scalable pipelines, partnering with researchers and platform

Qualifications

• Bachelor’s degree or equivalent experience in ML, data science, or related field.
• 6+ years in machine learning engineering or evaluation.
• Strong Python skills and experience building maintainable tools and data pipelines.
• Ability to translate evaluation questions into testable experiments.
• Excellent communication of findings to researchers and executives.

Responsibilities

• Design and maintain safety benchmarks for large language and multimodal models.
• Define metrics and protocols for safety and usefulness, with baseline comparisons.
• Build reusable Python pipelines with versioned datasets and experiment tracking.
• Develop automated red-teaming tools and validate scoring with human review.
• Investigate failures with researchers and create regression tests.
• Communicate results, limitations, and risks to stakeholders.

Skills

Python

Data pipelines

Experiment tracking

Statistics

Communication

Machine learning

Education

Bachelor's degree or equivalent experience

Job description

NVIDIA Corporation is seeking a Machine Learning Engineer / AI Researcher / Data Scientist to join the AI Safety and Responsibility efforts within Enterprise Risk Management. You will create robust benchmarks, run rigorous experiments, and build tools to evaluate safety of multimodal models, contributing to training methods and safeguards with reproducible results.

You will design benchmarks, define evaluation metrics, and implement scalable pipelines, partnering with researchers and platform