Research Engineer (Senior Staff+), Safeguards Labs

Research Engineers to define and execute AI safety research agenda.

• •Anthropic is seeking Research Engineers for its new Safeguards Labs team to investigate novel safety methods for AI systems.
• •You will define and execute research agendas, run experiments, and develop prototypes for safe model usage.
• •Key Responsibilities Lead research projects on detecting AI misuse and strengthening model safeguards.
• •Design and run analyses on model usage data to identify abuse patterns.
• •Develop and iterate on prototypes for real-time safeguards.
• •Contribute to research on detecting abusive behavior and training models to refrain from dangerous responses.
• •Build evaluations to measure safeguard effectiveness.
• •Requirements Track record of independently driving research projects from ambiguous problems to concrete results.
• •Comfortable scoping work and switching between research, engineering, and analysis.
• •Working familiarity with how large language models operate.