[GH-ISSUE #1498] DOC: <Issue related to /evaluation/how_to_guides/langgraph> #24

Open
opened 2026-02-17 17:19:00 -05:00 by yindo · 0 comments
Owner

Originally created by @JohnMunsch on GitHub (Feb 24, 2025).
Original GitHub issue: https://github.com/langchain-ai/docs/issues/1498

Wow. OK. We really wanted to put this code into use and believed that it was something ready for consumption given the article around it and the collected code at the bottom. It didn't take us long to find out that the collected code did not come close to running (not like, missing an API key, but missing chunks of code from up above).

Then when we approached it again to try and collect the code from the article itself, we again quickly realized that the code had been pulled from some kind of notebook (ala. Colab or Jupyter) and had problems with multiple assignments to the same variables, await calls at the top level (JavaScript/TypeScript is cool with that but Python is definitely not). Presumably most notebooks wrap each cell in some kind of async function and aren't freaked out by awaits in the cells.

Anyway, this was definitely not ready for public consumption. It either needed to be presented as what it was, a group of cells in a notebook, or it needed to be given to some newbie to run and make sure it was working runnable code (or that they could make it be that way as soon as they provided an API key or some kind of connection to a LLM).

P.S. My apologies for using a personal GitHub account to post this, our work GitHub accounts consider posting an issue on another company's software to be "contributing" and we are restricted from doing so.

Originally created by @JohnMunsch on GitHub (Feb 24, 2025). Original GitHub issue: https://github.com/langchain-ai/docs/issues/1498 Wow. OK. We really wanted to put this code into use and believed that it was something ready for consumption given the article around it and the collected code at the bottom. It didn't take us long to find out that the collected code did not come close to running (not like, missing an API key, but missing chunks of code from up above). Then when we approached it again to try and collect the code from the article itself, we again quickly realized that the code had been pulled from some kind of notebook (ala. Colab or Jupyter) and had problems with multiple assignments to the same variables, await calls at the top level (JavaScript/TypeScript is cool with that but Python is definitely not). Presumably most notebooks wrap each cell in some kind of async function and aren't freaked out by awaits in the cells. Anyway, this was definitely not ready for public consumption. It either needed to be presented as what it was, a group of cells in a notebook, or it needed to be given to some newbie to run and make sure it was working runnable code (or that they could make it be that way as soon as they provided an API key or some kind of connection to a LLM). P.S. My apologies for using a personal GitHub account to post this, our work GitHub accounts consider posting an issue on another company's software to be "contributing" and we are restricted from doing so.
yindo added the langsmithexternal labels 2026-02-17 17:19:00 -05:00
yindo changed title from DOC: <Issue related to /evaluation/how_to_guides/langgraph> to [GH-ISSUE #1498] DOC: <Issue related to /evaluation/how_to_guides/langgraph> 2026-06-05 17:24:37 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/docs#24