In this webinar, we’re excited to host Omar Khattab and Thomas Joshi to present DSPy – a framework for LLMs that emphasizes programming over prompting.
Instead of hand-crafting prompts for LLMs, with DSPy you can compose modules together in a declarative, Pythonic syntax (instead of manually writing prompts). You can then feed it to a compiler which will auto-tune the prompt (or even fine-tune the LM) to optimize the task for your use case!
Lots of cool concepts in here, many of which haven’t been explored in LlamaIndex yet.
Add comment