LGraph's persistence layer is a total game-changer for human-in-the-loop interactions. I'm loving the Interrupt tool - seamless input and resumption of nodes is huge.
@jingqiwu2865Ай бұрын
Awesome! been waiting for this solution for months. We have done some workaround to achieve the similar goal, but now it is much elegant.
@ki-werkstattАй бұрын
This is great! Always wanted to stop for a human input in an easy way. THX
@CharlotteLopez-n3iАй бұрын
I'm loving the new interrupt tool in L Graph - game-changer for human-in-the-loop interactions. Can't wait to dive into the use cases.
@RaviPrakash-dz9fmАй бұрын
Just what i needed. Was having to code this part myself for my app.
@kenchang3456Ай бұрын
Thanks for this feature. I agree, human in the loop is a good/necessary thing.
@KeyserTheRedBeardАй бұрын
Great video, LangChain. Looking forward to seeing your next upload from you. I pressed the thumbs up icon on your content. Keep up the amazing work! The introduction of the interrupt function seems like a game changer for human-in-the-loop systems. How do you see this impacting the future of multi-agent interactions, especially in terms of efficiency and oversight?
@NaveenReddy-p5jАй бұрын
Love the Interrupt feature! How do you see it impacting the development of multi-agent systems?
@AlexJohnson-g4nАй бұрын
Love the interrupt feature in L Graph! How do you see it improving human-in-the-loop interaction patterns in real-world agent development?
@aifarmerokayАй бұрын
Covers the streaming in langgraph how to do in production
@TrazynnАй бұрын
So far most videos have been on building multi-agent networks. But perhaps maybe take a step back and get more into something more basic and usable; which is automated prompt workflows. Scalable multi-prompt chains that aren't agentic but simply allow a team to process large amounts of qualitative data (i'm thinking of phone transcripts, chatlogs, customer interviews). Stuff that needs a chain of many prompts such that the context-windwos don't degrade the material.