Hello, I would like guidance on setting up a conversational loop using Crew AI’s kickoff command to make interactions feel continuous and responsive, like a chatbot conversation. Could you provide detailed steps or resources on achieving this?
I wrote a blog post on this topic so shameless plug here https://www.zinyando.com/building-a-conversational-chatbot-with-crewai-groq-chromadb-and-mem0/
You might want to add conversation history so that your chatbot remembers the current conversation. Feel free to ask if you have any questions.
Hello folks,
I am working on making a conversational chatbot using open source frameworks like ollama , langchain , langgraph and langchain based agents and tools but the agents don’t seem to be that much responsive using ollama with langchain , Because of which I want to use crew ai to build a conversational chatbot. But due to mem0 being a paid service I cant use it in my project. Do we have any other option to use instead of mem0 to maintain conversation history?
Mem0 has an Open Source option that you can host yourself and don’t have to pay. Have you tried that? Guide - Mem0.ai
Also, Mem0 doesn’t save conversation history but it saves long term memories about the conversation. You might need to save short term memory another way.
Hello friend,
Thanks for your reply. Can you suggest any method to store conversation memory.
You can keep conversation history in memory or store it in a database then only load an small portion of it eg last 50 messages as history in your conversation. For in memory you can use something like deque from collections. You can see the pattern in this blog post and you can port the concepts over to your conversational crew How to Add Long-Term Memory to AI Companions: A Step-by-Step Guide
Thanks, I will look into it.
@Toshi_Tyagi you can check this out: https://compextai.dev/
This should help you out. I’m building this conversation management open source tool for managing conversations better