Insights from Pydantic for LLM Workflows - Deeplearning.ai
Thanks to the course on Pydantic for LLM Workflows course. I learnt a lot in a short time. I had previous experience and some use of validation of json being passed to LLMs by using Pydantic to improve the schema validation autoamtion was another level of learning. I would recommend to take this course. The application extends to many use cases and the imagination is the only limit. Some of the ones I can think of is data schema validation and automating and self-servicing data engineering request changes being one of them.
Insights gained:
- Pydantic library for input validation and feedback looping through LLMs to correct errors and get the output in the right format.
- This is also helpful to automate system to system calls which have a input contract.
- Trick to improve the prompt through self-analysis and passing the model schema as hint.
- Use of instructor library as an abstraction to use to call any llm provider in the backend with a consistent api and responding with pydantic model output.
- This can be extended through use of Pydantic AI library which helps to use agentic ai with schema validation on top of pydantic reducing the need to write feedback logic.
Below you can find a snapshot of my course completion certificate.
LEAVE A COMMENT
Comments are powered by Utterances. A free GitHub account is required. Comments are moderated. Be respectful. No swearing or inflammatory language. No spam.
I reserve the right to delete any inappropriate comments. All comments for all pages can be viewed and searched online here. To edit or delete your comment: click the "Comments" link at the top of the comments section below where it says how many comments have been left (this will take you to a GitHub page with all comments for this page) --> find your comment on this GitHub page and click the 3 dots in the top-right --> click "Edit" or "Delete". Editing or adding a comment from the GitHub page also gives you a nicer editor.