Responsibilities
- Create datasets and moderation models to assess large language models and full systems for content safety and machine learning fairness, including text-only and multimodal inputs.
- Build training datasets for large language models using supervised fine-tuning and reinforcement learning methods across safety, fairness, and security domains.
- Investigate and apply state-of-the-art approaches to identify and reduce bias in language models and AI systems.
- Establish and monitor key performance indicators related to responsible use and behavior of language models.
- Adopt best practices in automation, system monitoring, scalability, and safety protocols.
- Enhance internal repositories and build safety-focused tools to support machine learning teams.
- Perform data cleaning, preprocessing, and transformation in collaboration with data scientists and engineers using large and diverse datasets.
- Carry out exploratory data analysis to detect patterns and insights that improve model effectiveness.
- Work with cross-functional teams including product engineers, data scientists, and analysts to convert business needs into machine learning solutions.
Compensation
Full benefits
Work Arrangement
Hybrid — Santa Clara, CA
Other
- Full-time (W-2) contract role
- Full benefits
- PTO
