LLM training data & evaluation
Generative-AI workflows to produce precise, reviewable datasets that improve model behavior—experience from Outlier collaborating on OpenAI-facing programs.
•
1 min read
Generative-AI workflows to produce precise, reviewable datasets that improve model behavior—experience from Outlier collaborating on OpenAI-facing programs.
Convolutional policy trained with deep reinforcement learning in a custom Python simulator—team lead for a summer DISCOVERY LAB GLOBAL internship.