Machine Learning Engineer - Reward-modeling Specialist

Company Information: Encultured AI is a for-profit video game company with a public benefit mission: to develop technologies promoting the long-term survival and flourishing of humanity and other sentient life. We aim to this by creating a game for fun, a laboratory for safety, and a sandbox for culture, all on one platform. You can learn more about the company on our website: Position Description: In this role you would train, test, and deploy reward-modeling algorithms for aligning with and assisting human players in a gameplay setting. You would work closely with a team of fellow machine learning engineers to reach development milestones. Location: You would primarily work remotely, with regular Zoom calls and Discord communication. At least once per week we work together in person in the San Francisco / Berkeley area for an in-person workday, so if you live nearby it’d be great to have you attend those. Compensation: Starting salary would be between $120k and $180k per year depending on experience, plus healthcare benefits, and equity incentives vesting over 5 years, with raises also becoming available with good individual performance or team-wide accomplishments that expand our revenue stream. We also offer a Safe-Harbor 401(k) with the IRS maximum employer matching. Qualifications: In this role we need candidates to have experience with applying and developing algorithms that learn to model human preferences represented as reward functions. In addition, we have two tiers of qualification for this role: Junior ML Engineer and Senior ML Engineer. To qualify directly for the Senior title upon joining, we require candidates who have received a Bachelor’s degree or above in computer science, physics, mathematics, or a closely related field. These are not required, but our team will welcome and make good use of experience with: -building and training reinforcement learning algorithms and/or environments, -building and training large language models (LLMs), -applying and developing AI alignment methods, -research on humans and human culture, including but not limited to background in the humanities, social sciences, cognitive science, and biology, -machine learning interpretability research and tools, -PhD-level research and writing.
Encultured AI is looking to hire a Machine Learning Engineer - Reward-modeling Specialist at their Remote location. Job contains the tags Remote, Physics, AI, English, Environment and Character.
Job discovered on 8/22/2022