Applied Safety Research Engineer, Safeguards
Is deze vacature iets voor u?
Mijn cv maken Maak uw cv en ontdek uw matchpercentage met deze functie — en met alle andere.
De functie
Design and run experiments to improve evaluation quality, including methods to generate representative test data, simulate realistic user behavior, and validate grading accuracy.
Research how factors such as multi-turn conversations, tools, long context, and user diversity impact model safety behavior.
Analyze evaluation coverage to identify gaps and determine where better measurement is needed.
Productionize successful research into evaluation pipelines that run during model training, launch, and beyond.
Collaborate with Policy and Enforcement to translate real-world harm patterns into measurable evaluations.
Build tooling that enables policy experts to create and iterate on evaluations, and surface findings to drive upstream model improvements.
Research how factors such as multi-turn conversations, tools, long context, and user diversity impact model safety behavior.
Analyze evaluation coverage to identify gaps and determine where better measurement is needed.
Productionize successful research into evaluation pipelines that run during model training, launch, and beyond.
Collaborate with Policy and Enforcement to translate real-world harm patterns into measurable evaluations.
Build tooling that enables policy experts to create and iterate on evaluations, and surface findings to drive upstream model improvements.
Bekijk de volledige vacature
Taken, profiel, vaardigheden en voordelen — maak gratis een account aan.
of
Al een account?
InloggenVergelijkbare vacatures
Andere functies die kunnen passen.