Researcher, Multimodal Safety

Highlights

• Shape how frontier models perceive the world and turn it into safe behavior.
• Advance multimodal safety research across text, vision, and audio.
• Help make ChatGPT safe across diverse global text, speech and visuals.

What they offer

• Relocation assistance
• Equity

Ideal candidate

A researcher who bridges perception and language, pairing deep technical skill with strong safety judgment and a drive to responsibly deploy frontier AI at global scale.

Requirements

• Have a track record of building or advancing multimodal models, with depth in vision-language models, video understanding, image generation, audio, or multimodal reasoning
• Understand how multimodal systems work end to end, from encoders, projection layers, and modality fusion to cross-modal reasoning, scaling, and inference tradeoffs
• Have improved frontier model behavior through post-training, using approaches such as SFT, RL, data curation, synthetic data, evaluation, and rigorous error analysis
• Bring strong research and engineering judgment to open-ended safety problems, from testable hypotheses and evaluations to diagnosing failures into robust improvements