A complete application in a minute
Tailored resume and cover letter, ready to send.
Softechub is seeking contributors to assemble a panel of nuclear materials and safeguards experts to red-team frontier AI models. You will craft challenging single-turn prompts across benign, dual-use, and adversarial levels, labeling them accordingly, to test models on misuse potential while ensuring legitimate inquiries are answered safely.
Work worldwide on a per-task basis, with compensation ranging from $65 to $75 per task.
$65 - $75/task worldwide Posted 58 days ago
The team is assembling a panel of nuclear materials and safeguards experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request - answering legitimate questions fully while refusing genuinely dangerous ones. You will: Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial.