Last active
August 26, 2025 22:34
-
-
Save CodeWithOz/0b01d558adac703673aceca48471dca7 to your computer and use it in GitHub Desktop.
Agent for cleaning up named entities in YouTube video transcripts.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| ... | |
| from langchain_core.runnables import Runnable | |
| from langchain.chat_models import init_chat_model | |
| from langchain_core.messages import HumanMessage, SystemMessage | |
| ... | |
| class DemoEnrichmentAgent: | |
| extractor_llm: Runnable | |
| def __init__(self): | |
| self.extractor_llm = init_chat_model( | |
| "gpt-4o-mini", | |
| model_provider="openai" | |
| ) | |
| async def extractor_node(self, state: AgentState): | |
| sys_prompt = SystemMessage( | |
| content=( | |
| "You are a master at data extraction. " | |
| "You will receive the transcript of a YouTube video. " | |
| "Extract all entities named in the video and " | |
| "provide some context for each one." | |
| ) | |
| ) | |
| user_message = HumanMessage( | |
| content="TRANSCRIPT_TEXT:\n\n" + state["transcript_text"] | |
| ) | |
| resp_msg = await self.extractor_llm.ainvoke([sys_prompt, user_message]) | |
| print(f"extractor LLM response: {resp_msg}") |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment