QVAC SDK 0.9.0 enables LoRA fine-tuning on-device for LLM customization without cloud data sharing, releasing in 10 days from Tether's Paolo Ardoino.

According to DeepLearning.AI, its Post-training of LLMs course teaches how to customize pre-trained language models using Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning (RL) (source: DeepLearning.AI on X, Oct 6, 2025). According to DeepLearning.AI, the curriculum explains when to use each method, how to curate training data, and how to implement the techniques in code to shape model behavior effectively (source: DeepLearning.AI on X, Oct 6, 2025). According to DeepLearning.AI, enrollment is available via the provided link hubs.la/Q03MrTZS0 (source: DeepLearning.AI on X, Oct 6, 2025).