Reinforcement Fine-Tuning in Production
Get expert answers on RFT, GRPO, and reward design.
This webinars is hosted by AI experts behind the course “Reinforcement Fine-Tuning LLMs with GRPO” for Deeplearning.AI!
Reinforcement Fine-Tuning (RFT) is no longer experimental. It’s powering real-world GenAI systems with tighter feedback loops and better reasoning.
But how do you actually make it work in production?
Watch Let’s Talk Tokens, a 1-hour AMA with the engineers behind the definitive RFT course from DeepLearning.AI.
What we’ll cover:
- When RFT outperforms supervised fine-tuning (and when it doesn’t)
- How GRPO enables scalable reward-based training
- Reward design strategies that avoid mode collapse & reward hacking
- Best practices for evaluating RFT-trained models in production
If you’re fine-tuning models or planning to, this session is built for you.
No fluff. Just real answers from real engineers who’ve done it.