The OpenAI staff posted
See section
Evaluation, tuning, and shipping safely
- Evals API for eval-driven development.
- Reinforcement fine-tuning (RFT) using programmable graders.
- Supervised fine-tuning / distillation for pushing quality down into smaller, cheaper models once you’ve validated a task with a larger one.
- Graders and the Prompt optimizer helped teams run a tighter “eval → improve → re-eval” loop.
Be sure to give this post a like if you find the info of value so the Open AI staff know