Intro to LLM Agents – a economic industry data chat agent using OpenBB and LangChain



00:00 Intro
01:10 OpenBB overview
02:15 OpenBB demo
06:40 Agents, what they are, prompt and tools
12:15 Demo the Streamlit app
18:00 Advanced agent workflows
23:37 Limitation of LLMs and Agents
30:05 Wrap-up

My online journal post https://druce.ai/2024/05/OpenBB-LangChain-agent

Two excellent online journal posts about LLM agents
https://interconnected.org/home/2023/03/16/singularity
https://til.simonwillison.net/llms/python-react-pattern

Centaurs and Cyborgs paper / BCG study
https://www.oneusefulthing.org/p/centaurs-and-cyborgs-on-the-jagged
https://d3.harvard.edu/navigating-the-jagged-technological-frontier/

Code –
– top level Streamlit app – https://github.com/druce/openbb_agent/blob/main/streamlit_bbagent.py
– system prompt – https://github.com/druce/openbb_agent/blob/main/ainb_const.py
– tool definitions – https://github.com/druce/openbb_agent/blob/main/obbtools.yaml
– tools – https://github.com/druce/openbb_agent/blob/main/bb_tools.py

OpenAI and LangChain agent APIs
https://platform.openai.com/docs/guides/function-calling
https://python.langchain.com/v0.1/docs/modules/tools/


Disclaimer:

If you own the YouTube channel related to this video and do not want it to be featured here, you can contact us through our contact page. We will gladly remove it without questioning your reasons.

Leave a Comment