reward hacking Jobs

Abundant is an applied research lab focused on scaling reinforcement learning for safe and reliable agentic capabilities. As a Member of Technical Staff, Alignment, you will be instrumental in ensuring that alignment is a core component of all our work, from the code we deploy and the runbooks we write to the papers we publish. This role involves testing models in advanced environments, identifying and mitigating potential risks like reward hacking and deception, and building robust tools to monitor and secure AI agents.

$250k - $450k