Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Remotasks
Scale AINo subject matter listed
Image
TextI developed rubrics for assessing the accuracy and completeness of the responses by four different models, while providing detailed feedback to project managers and junior annotators to improve data quality. Additionally, I conducted research for robust justifications for penalizing unsafe responses, including identifying “jailbreak” conversations. Finally, I tested AI models by iteratively increasing prompt complexity, achieving over a 30% improvement. I also engineered multi-turn prompts with category, persona, semantic, and formatting constraints.
Writing prompts, rating models answers, developing rubrics for training the models, and re-writing the responses. Editing other annotators' work.
Writing prompts to generate model failure, provide LIKERT ranking and re-write the responses.
Annotating client-generated prompts in English, filtering out jailbreaks and harmful prompts.
Specialist Bootcamp, AI And Data Science
Master of Philosophy, Development Studies
AI Expert
RLHF annotator/reviewer
