SD Solutions

1 open remote positions

This company builds specialized evaluation infrastructure for AI safety and robustness testing. Their platform simulates adversarial conditions to validate AI agent behavior before deployment, and they are currently expanding their contributor network.

Open Positions

Vulnerability Researcher

Poland Bulgaria

  • Review cybersecurity prompts submitted to an AI system that cannot be confidently classified as benign or malicious.
  • Determine the underlying intent behind each prompt and assess whether the user is pursuing a legitimate purpose or intent to cause harm.
  • Annotate each prompt with an intent label and a clear rationale, escalating ambiguous cases to a reviewer.

View details Similar jobs