Job Summary
The position involves the development of reward models and evaluative criteria essential for ensuring the reliability and integrity of large language models (LLMs) throughout their training process. The individual will perform alignment experiments and create safety assessments that are applicable in practical environments.
Responsibilities
- Develop and implement reward models to enhance LLM performance.
- Conduct alignment experiments to assess model behavior and accuracy.
- Design effective safety evaluations relevant to real-world applications.
- Communicate complex findings regarding alignment in a clear and understandable manner.
Qualifications
- Demonstrated experience in training or evaluating large language models.
- Expertise in reward modeling, reinforcement learning with human feedback (RLHF), or preference learning.
- Strong programming skills in Python and experience with PyTorch.
- Ability to effectively convey intricate alignment results to diverse audiences.
ellamind provides equal employment opportunities to all employees and applicants for employment and prohibits discrimination and harassment of any type without regard to race, color, religion, age, sex, national origin, disability status, genetics, protected veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by federal, state or local laws.
