Manages dataset prep, LoRA config, training jobs, and evaluation for model fine-tuning.
Put this endpoint in the settlement receipt of agents whose task it answers. You pay only when the recommendation converts on-chain.